Interpretative applications of artificial intelligence in musculoskeletal imaging: concepts, current practice, and future directions

Abstract

Abstract: Artificial intelligence (AI) promises wide-reaching impacts on the field of radiology, and has the potential to influence every aspect of image interpretation. In recent decades, significant advancements in computing power, combined with the availability of large data stores or “Big Data” and algorithm democratization have revolutionized AI and machine learning (ML). Research applications utilizing these technological advancements are booming, and their adoption is expected to continue to rise at a rapid pace. While AI and ML have impacted many components of the imaging value chain, the purpose of this article is to discuss interpretative uses of the technology as it relates to musculoskeletal (MSK) radiology. This review provides a general introduction to AI and ML concepts, and highlights the major promises, challenges, and anticipated future applications of these developments in MSK radiology. AI and ML advances for image interpretation can increase the value that MSK radiologists provide to their patients, referring clinicians, and organizations by increasing diagnostic accuracy while decreasing turnaround times, enhancing image processing and quantitative analysis, and by potentially improving patient outcomes. Familiarity with these processes among MSK clinicians and researchers will be paramount to the improvement and implementation of these new techniques into the clinical practice. Radiology departments, practices and practitioners who embrace these technologies now will be well-suited to lead this influential change in our field in the near future.

Ähnliche Arbeiten