Neurosurgery does not stop teaching us when training ends, and neurosurgeons never stop learning. In a new five-part series, Graham McMahon, MD, MMSc, President and CEO of the Accreditation Council for Continuing Medical Education (ACCME), examines how experienced surgeons continue to learn, reflect and cultivate mastery throughout their careers.
In this first article, McMahon examines how artificial intelligence can help neurosurgeons navigate an ever-growing body of knowledge, identify gaps in their understanding and develop more focused learning around real clinical questions.
Be sure to listen to the companion Neurosurgery Podcast episode featuring McMahon, where he expands on the ideas behind the series.
Article 1
Dr. Chen had always taken pride in staying current. Journals arrived regularly in his inbox, and he attended at least one major neurosurgical meeting every year. Yet when a rare spinal tumor appeared on his clinic schedule one morning, he realized how difficult it had become to keep up with the pace of change in the literature.
Within minutes, he opened an AI assistant and asked a series of questions. What were the most recent developments in management of this tumor type? Were there new data comparing operative approaches? What were the key controversies emerging in the field?
The responses were not definitive answers. But they helped him quickly orient himself, identifying key studies, areas of disagreement and several questions worth exploring more deeply before seeing the patient. What he had discovered was not a shortcut to thinking. It was something more useful. A new way to organize his learning.
For most of modern medicine, the challenge for clinicians has been access to information. Today the challenge is different. Neurosurgeons are surrounded by more information than ever before. New devices, evolving techniques, expanding evidence and ever-growing literature compete for attention.
The real difficulty is not finding information. It is deciding what matters for the decisions we face in practice.
Artificial intelligence tools may offer a surprisingly useful way to navigate this challenge. Used thoughtfully, they can function less as authorities and more as learning partners, helping clinicians identify gaps in their understanding, explore emerging evidence and build targeted learning pathways around real clinical questions.
One of the most practical roles for AI is as a content curator. When faced with an unfamiliar or evolving topic, a surgeon can quickly ask an AI tool to summarize recent developments, identify key trials or outline areas of ongoing debate. Rather than spending hours sorting through search results, the clinician can rapidly develop an overview of the landscape before diving deeper into primary sources. The immediacy of content aligned with a patient’s needs makes learning especially powerful.
AI can also function as a learning pathway engineer. The most effective learning for experienced clinicians is often driven by real clinical questions. Preparing for an unfamiliar case, examining an unusual complication or revisiting a difficult decision can all become opportunities for focused learning.
AI tools can help generate reading lists, highlight contrasting viewpoints and map out the key considerations surrounding a clinical problem.
Of course, these tools must be used carefully. AI systems can produce confident but inaccurate statements, compress nuance in the literature or rely on outdated information. The decisiveness of the recommendation can feel seductive. However, to minimize the risk of deskilling, these outputs are best used as starting points for exploration rather than definitive authorities. Responsible clinicians will continue to verify important claims through primary sources and established guidelines.
When used in this way, AI can help transform the rhythm of professional learning. Instead of relying primarily on periodic conferences or journal reading, neurosurgeons can develop a form of just-in-time education that responds directly to the questions arising from their own patients and cases.
The next time you prepare for an unfamiliar case, try asking an AI assistant three questions: What has changed recently in the literature? What are the major areas of disagreement? What questions should I be asking before I see this patient?
The surgeons who benefit most from these tools will not be those looking for quick answers. They will be those who use them to ask better questions.
Reference
Chew KS, Durning SJ, van Merrienboer JJG. Teaching metacognition in clinical decision making using a checklist approach. Medical Education. 2016. https://pmc.ncbi.nlm.nih.gov/articles/PMC5165179/
Graham McMahon, MD, MMSc
Graham McMahon, MD, MMSc, is the president and CEO at the Accreditation Council for Continuing Medical Education (ACCME) based in Chicago. Previously he was an associate dean and associate professor of medicine at Harvard Medical School, and program director for the fellowship program in the division of endocrinology, diabetes and hypertension at the Brigham & Women’s Hospital in Boston where he completed his postgraduate training.
