Implementing AI in breast imaging: challenges to turn the gadget into gain

Excerpt

In healthcare, the integration of artificial intelligence (AI) has gained significant traction. In regard to the breast imaging profession in particular, there is increasing pressure due to the exponential growth in imaging services in tandem with a predicted reduced workforce to provide these services. Correspondingly, there is growing interest within the breast imaging community to adopt AI to optimize workflow efficiency and increase productivity, and concurrently improve patient outcomes. While the adoption of broad AI in breast imaging clinical practice has been more or less a silent revolution, with AI nowadays being used in multiple areas such as invite coordination, appointment scheduling, report generation, image positioning, and quality insurance, the implementation of narrow AI is not happening as rapidly [1‐3]. The issue is not quite the availability of AI products; there are multiple commercially available AI products for breast imaging that claim to improve efficiency and quality or care (https://​grand-challenge.​org/​aiforradiology/​breast), but their adoption is not ubiquitous, due to several possible influencing factors. …