The BUS-BRA Dataset is a publicly available dataset of anonymized breast ultrasound (BUS) images of 1,064 patients that underwent routinary breast studies. The BUS dataset contains biopsy-proven tumor cases and BI-RADS annotations in categories 2, 3, 4, and 5. In addition, the dataset also contains ground truth delineations that divide the BUS images into tumoral and normal regions. A full description of the BUS-BRA dataset can be found in the following article: Wilfrido Gómez-Flores, Maria Julia Gregorio-Calas, and Wagner Coelho de Albuquerque Pereira, "BUS-BRA: A Breast Ultrasound Dataset for Assessing Computer-aided Diagnosis Systems," Medical Physics, vol. 51, pp. 3110-3123, 2024, DOI: 10.1002/mp.16812. Any research originating from its usage is required to cite this paper. The Program of Biomedical Engineering of the Federal University of Rio de Janeiro (PEB/COPPE-UFRJ, Brazil) has a copyright on the data and is the principal distributor of the BUS-BRA Dataset. In addition, the Centro de Investigación y de Estudios Avanzados (Cinvestav, Mexico) is involved in an ongoing effort to develop this dataset to aid research efforts in developing, testing, and evaluating algorithms for breast ultrasound analysis.

The source codes to replicate these experiments in the article are found at https://github.com/wgomezf/BUS-BRA