Machine-learning for osteoarthritis research

Machine learning (ML) algorithms have the ability to automatically learn and improve from experience without being specifically directed. There has been great optimism that such techniques may improve scientific biomedical research in several fields1. Although conventional statistical modeling remains the method of choice for etiology-driven and explanatory analyses, ML consists in an interesting approach to identify new associations and patterns in large datasets. This is particularly important now, as it has become possible to generate large quantities of data from each study participant, from sources such as high-resolution MRI imaging, serum sample analysis, genome sequencing, and electronic medical records