Date of Award
Master of Medical Science (MMSc)
Reni Butler, MD
Triple-negative breast cancer is a subtype of invasive ductal carcinoma that lacks expression of estrogen, progesterone, and human epidermal growth factor 2 receptors. Triple-negative cancers are associated with poorer prognosis partly due to their benign imaging features, leading to possible delay in diagnosis. Computer-aided diagnosis, which provides radiologists with a preliminary assessment of a breast lesion based on artificial intelligence training, is a proposed technique that may improve cancer detection with ultrasound. However, very few studies investigated its use in the diagnosis of triple-negative breast cancers. In this study, we will assess the efficacy of computer-aided diagnosis in improving the diagnostic accuracy of ultrasound for triple-negative breast cancers. Using a blinded crossover retrospective review, we will compare reader accuracy of radiologists evaluating triple-negative breast cancers with computer-aided diagnosis versus standard unaided interpretation. This study may provide support for using this technology in clinical practice to help diagnose triple-negative breast cancers.
Burick, Isabel, "Computer Aided Diagnosis with Ultrasound to Improve Detection of Triple Negative Breast Cancer" (2021). Yale School of Medicine Physician Associate Program Theses. 93.