Breast Series: Emerging Technologies (2023)-RC31519-2023

Video Transcription

Video Summary

Breast cancer imaging has evolved significantly with technology, particularly in the realm of computer-aided detection (CAD) and diagnosis, affecting how radiologists perform screenings. Initially, CAD served as a secondary tool for radiologists, marking potential abnormalities on imaging; however, technology advances have shifted its role to a concurrent reader in screening processes. CAD assists in evaluating dense breast tissues, especially in 3D imaging contexts such as whole breast ultrasound and dynamic contrast-enhanced MRI.<br /><br />CAD systems have been developed to potentially enhance radiologist accuracy and efficiency, a necessity given the increasing volume of 3D imaging data. Studies show CAD's ability to reduce image interpretation time without sacrificing diagnostic accuracy, vital for screening programs with limited resources. Emerging equities like AI-based synthetic 2D images generated from 3D tomosynthesis stacks add supplementary diagnostic value, assisting in overcoming tissue superposition challenges in breast imaging.<br /><br />The conversation around CAD evolution includes independent reading, where AI alone might identify normal cases for triage, thereby reducing radiologists' workloads without impacting sensitivity or specificity negatively. Converging CAD and AI further impacts patient management by refining imaging protocols and workflow efficiencies, emphasizing a decrease in false-positive rates while maintaining high diagnostic standards.<br /><br />Beyond CAD for detection and diagnosis, AI contributes to multi-omics cancer discovery through imaging, assessing tumor features like heterogeneity linked to genetic expressions, suggesting new pathways for precision medicine. Continuous learning from diverse data sets, addressing time interval disparities, and integrating biologically relevant features into predictive models remain crucial objectives in evolving imaging landscapes. Emphasis lies in AI's role as an aid, not a replacement, ensuring adherence to clinically safe and effective practices while advancing clinical outcomes and patient care delivery.

Keywords

breast cancer imaging

computer-aided detection

radiologists

screenings

dense breast tissues

3D imaging

whole breast ultrasound

dynamic contrast-enhanced MRI

AI-based synthetic images

tissue superposition

independent reading

AI triage

multi-omics cancer discovery

precision medicine

false-positive rates