Artificial Intelligence in Shoulder Arthroplasty

Artificial intelligence (AI) and machine learning have thus far had only modest influence in the field of shoulder arthroplasty, but AI will soon be increasingly used to create and study “big data.” There is tremendous optimism and hope regarding the ways in which AI could benefit the field of shoulder arthroplasty. However, while AI holds great potential for evaluating and assessing large datasets and information unable to be processed by the human mind, the ability of computers to “learn” and provide accurate and clinically useful information for total shoulder arthroplasty has not yet been realized. Evaluation and management of large datasets by AI can decrease time spent and increase data accuracy, but AI is inherently dependent upon what is reported in the literature so scientific studies published with incomplete data or short follow-up intervals can lead to conclusions that may be inaccurate or expressed only as percentages. This chapter will assess the progress made in AI for effective image recognition, determination of risk of various treatments, determinants of case outcomes, and clinical decision-making.