Background: Breast Cancer incidence in the United States is expected to increase by almost 50% over the next few decades. Artificial intelligence (AI) offers hope to combat this trend, with rapid acceleration and promising clinical advances. In Radiology, breast imaging has been at the forefront of this innovation, particularly in detection. However, AI has also shown promise in improving the patient journey through the breast imaging center and beyond, including in the areas of engagement and education through generative AI, intake, imaging acquisition, diagnosis, reporting, results communication, and follow-up care.
Learning Objectives: • Understand the current state of AI, machine learning, and deep learning in breast imaging and project the future practice of breast imaging • Identify the key limitations of AI in breast imaging. • Recognize the potential benefits of incorporating AI into a breast imaging practice beyond detection and diagnosis, including all aspects of the patient journey. • Understand the challenges that need to be addressed before AI can be implemented into daily clinical practice for breast radiologists • Review regulatory considerations of AI in clinical practice
Abstract Content/Results: In this abstract, we will discuss the current state of AI in breast imaging and its potential future impact on clinical practice and the diagnosis of breast cancer. We will review the key limitations and potential hurdles of AI in breast imaging including limited FDA approvals, limited datasets with ground truth for training algorithms, bias, issues with data privacy and ethical considerations. Finally, through a series of patient personas from different demographics and breast cancer risk profiles, we will demonstrate AI innovations and technologies under current development in future clinical practice to help illustrate how breast cancer may be detected in the next decade.
Conclusion: With breast cancer incidence on the rise, the future will depend on breast imagers leveraging innovations to deliver precision care at scale. AI has the potential to revolutionize breast imaging by improving patient engagement and education, boosting quality, accuracy, and efficiency, and detecting cancer even earlier than thought possible (potentially before detectable on today’s imaging modalities). Ultimately, thoughtful utilization of these tools will help drive patient compliance and adherence to screening mammography and supplemental screening modalities.
