Artificial Intelligence (AI) can be used to detect lesions and discover new imaging biomarkers for early diagnosis and prevention of diseases. In this way, more precise disease monitoring and follow-up is enabled, providing an extra layer of image interpretation and useful information.

Frequently, in medical images obtained for other diagnostic questions, incidental findings are present. Despite being detectable in the images, they are often not picked up due to their subtle appearance and foremost time constraints in the clinical practice. Preventive early detection of disease can be achieved when the power of AI is used as an extra pair of eyes to support radiologists in the background.

AI is being used in radiology in a number of ways today. However, deep learning algorithms have not yet been applied in real-world imaging into applicable use cases in imaging of structural joint damage caused by inflammation in rheumatic diseases. For the most effective use of AI, it is necessary to coordinate the input of high-quality expert-labelled medical images and algorithm development between physicians and computer scientists. These use cases can solve increasingly challenging problems in clinical practice. This (r)evolution in imaging depends on applying powerful AI tools to clinical problems in the best possible ways, i.e., the successful development of AI use cases.