Impact of Machine Learning in Neurosurgery: A Systematic Review of Related Literature

Abstract

Machine learning is a domain within artificial intelligence that allows for computer algorithms to be learned from experience without them having being programmed. The objective of this study is to summarize the neurosurgical applications of machine learning when compared to clinical expertise. This study uses a systematic search to review articles from the PubMed and Embase databases in comparing various machine learning studies approaches to that of the clinical experts. For this study, 23 studies were identified which used machine learning algorithms for the diagnosis, pre-surgical planning, and outcome prediction. In conclusion, this study identifies that machine learning models can augment decision-making capacity for the surgeons and clinicians in neurosurgical applications. Despite this, there still exist hurdles that involve creation, validation, and the deployment of the machine learning techniques in clinical settings.