J Korean Orthop Assoc.  2023 Dec;58(6):460-467. 10.4055/jkoa.2023.58.6.460 .

Clinical Research and Trends in Orthopedic Surgery and Musculoskeletal Tumor Using Artificial Intelligence

• Seo SW

• KMID: 2549820
• DOI: http://doi.org/10.4055/jkoa.2023.58.6.460

Abstract

Clinical cancer researchers have been studying means of predicting the prognosis of cancer and the effectiveness of treatment based on various patient data. The information that can be obtained from patients is diverse, including medical images, genetic data, bio-signals, text data, and biochemical tests. Although traditional epidemiological methodologies have been used for data analysis, recent studies have revealed the superior performance of machine learning algorithms. On the other hand, medical staff directly involved in utilizing artificial intelligent (AI) medical devices in the future will need to understand the strengths and limitations of these algorithms in detail. Machine learning algorithms do not work on a single principle, and various algorithms originating from various backgrounds are available. Each algorithm may or may not be suitable, depending on the purpose. Algorithms not correctly applied for this purpose often cannot be used clinically, even if their performance is excellent. This review aims to provide an overview of the clinical utility and constraints of AI medical devices by introducing various AI algorithms applicable to clinical practice and highlighting relevant case studies.

Keyword

machine learning; cancer; artificial intelligence; musculoskeletal tumor