Mapping the AI (ML) Chasm in Neurosurgical Oncology: A Scoping Review

Abstract (English)

Machine learning (ML) algorithms have played an important role in contemporary neurosurgical cancer research for more than a decade. For their high impact potential, many ML models have been developed to potentially improve preoperative, intraoperative, and postoperative neurosurgical care upon implementation. Nevertheless, though many ML models have been developed with the intent of improving neurosurgical oncology care, few studies have evaluated the efficacy and/or effectiveness of these algorithms in practice. This gap between ML model development and implementation is being increasingly referred to as the "AI chasm" in the ML literature . This chasm obscures whether ML has actually led to measurable improvements in neurosurgical oncology care. This scoping review aims to assess the prevalence of ML implementation research in neurosurgical oncology in the last two decades.