Applications, limitations of AI for fracture detection, classification

Editor's Note

Preliminary experience in the automated detection and classification of fractures using artificial intelligence (AI) shows promise, and AI may enhance processing and communicating probabilistic tasks in orthopedic surgery, this study finds.

For fracture detection, researchers compared the human findings in 10 studies with AI findings. In two studies, AI outperformed human examiners for detecting and classifying hip and proximal humerus fractures, and in one study, AI was equivalent to human detection of wrist, hand, and ankle fractures.

Presently, inadequate reference standard assignments to train and test AI is the biggest hurdle for integration into clinical workflow, the authors say. Future studies should address legal regulation and feasibility of implementation in clinical practice.