Artificial intelligence is currently transforming our everyday life. In myriads of ways we are using AI to make tasks easier or quicker. However, in the narrow application of medical imaging interpretation the use of AI is still very limited. This is likely to change in the coming years due to increasing quality of AI evaluation in standard situations, breast screening being among the forefront due to the very large numbers. Current high-quality AI applications, available for clinical use, show performance that meet, if not surpass that of dedicated breast radiologists. How to use these techniques to our advantage is however still an unanswered question. This is mainly due to limited data about the effects of the use of AI in real world settings. In this lecture the possibilities to overcome this final hurdle, and the potential pitfalls associated with use of AI will be discussed.
Learning Objectives:
To appreciate the quality of current AI tools for mammography and DBT
To be able to discuss the benefits and drawbacks of using AI for image interpretation
To understand the likely future roles of AI in the practice of breast radiologists
