Exploring the Current Applications of Artificial Intelligence in Orthopaedic Surgical Training: A Systematic Scoping Review

Abstract

In recent years, the integration of artificial intelligence (AI) in surgical education has been prominent, as evidenced by recent publications. Given the unique requirements and challenges associated with orthopaedic training, we conducted a systematic scoping review that examined the applications of AI only in this setting. A comprehensive search was conducted across four databases: Embase, CENTRAL, Medline, and Scopus. Original research articles that utilised an AI model within a specific orthopaedic educational context were considered for inclusion. Data from the included studies were extracted onto a bespoke form, followed by a thematic analysis to detect patterns within the data. Our findings were then summarised descriptively. A total of 21 studies were included in the scoping review, encompassing 273 participants. In relation to the integration of AI in orthopaedic surgical training, two overarching themes were identified: refinement of surgical competencies and enhancement of knowledge acquisition. All included studies, with the exception of one, were conducted in the last five years. Twelve distinct AI models were utilised across the included studies, with large language models accounting for over half the applications. Multiple promising interventions were highlighted, particularly the use of personalised automated feedback models for evaluating performance in surgical tasks. AI holds major potential to revolutionise orthopaedic surgical training. However, evidence supporting its use in this field remains limited. Further studies, preferably randomised controlled trials with larger sample sizes, are required to strengthen the evidence base.

Ähnliche Arbeiten