Artificial Intelligence in Breast Imaging (2023)-WEB40-2023

Video Transcription

Video Summary

The RSNA webinar on "Artificial Intelligence in Breast Imaging," led by Dr. Manisha Baha, featured discussions on the implementation and advancements of AI in breast cancer detection and diagnosis. Key speakers included Dr. Emily Conant, Dr. Despina Contos, and Dr. Sophia Zacherson, who highlighted the integration of AI algorithms in mammography and tomosynthesis to improve accuracy and efficiency in diagnosing breast cancer.<br /><br />Dr. Bahl introduced the session by explaining AI's role in radiology, emphasizing its subfields, such as machine learning and deep learning, which are pivotal in developing algorithms that enhance imaging diagnostics. She explained concepts like supervised versus unsupervised learning, crucial in AI development, and internal versus external validation for model performance evaluation.<br /><br />Dr. Zacherson provided insights from a European perspective, focusing on AI's capabilities like workload reduction and early cancer detection, supported by recent studies showing AI’s potential to replace one of the radiologists in double-reading screening programs. She stressed the importance of validating AI tools for diverse clinical and demographic environments to ensure effectiveness.<br /><br />Dr. Conant highlighted AI applications in tomosynthesis, especially in the U.S., discussing studies that reveal AI's ability to improve sensitivity, specificity, and reading efficiency. She also mentioned the potential of AI for risk assessment in breast cancer screening, leveraging advanced imaging data.<br /><br />Dr. Contos discussed the use of radiomics and new AI directions for better breast cancer risk assessment, noting that AI should incorporate diverse data sources for improved personalized medicine. The webinar concluded with a Q&A session addressing concerns about AI bias and the reliability of unsupervised learning. Overall, the webinar underscored AI's promising role in enhancing breast imaging, addressing ethical considerations, and fostering innovation.

Keywords

Artificial Intelligence

Breast Imaging

AI in Radiology

Breast Cancer Detection

Mammography

Tomosynthesis

Machine Learning

Deep Learning

Radiomics

AI Bias