Aims: This thesis thoroughly examines the impact of breast density and human-related factors on performance and the diagnostic efficacy of DBT and ultrasound as assessment tools. Methods: The first study recruited 128 readers to interpret DM test sets. Participants completed a survey about their backgrounds. Each reader interpreted the images, localised any lesions identified, stated the feature, and provided a confidence score. The second study recruited 11 readers to interpret DM and DBT test sets. The third study examined the diagnostic efficacy of DBT and ultrasound in 482 recalled women. The final study assessed the potential for DBT and ultrasound to reduce unnecessary biopsies in 550 recalled lesions. Results: The first study showed that participating in a screening program reading, breast imaging fellowship and weekly interpretation of >100 mammograms are the most important indicators of diagnostic performance. The second study showed that DBT significantly improved diagnostic confidence in dense (p=0.03) and non-dense breasts (p=0.003) when reporting breast lesions, but not in other diagnostic metrics. The third study showed that DBT sensitivity was significantly higher than ultrasound in dense breasts (p100 mammograms weekly can optimise cancer detection in dense breasts. Findings also indicate that DBT is an important tool for optimising breast cancer screening and assessment in women with dense breasts, and that ultrasound has greater potential for reducing unnecessary biopsies.
