BUS-BRA: A breast ultrasound dataset for assessing computer-aided diagnosis systems.

PURPOSE: Computer-aided diagnosis (CAD) systems on breast ultrasound (BUS) aim to increase the efficiency and effectiveness of breast screening, helping specialists to detect and classify breast lesions. CAD system development requires a set of annotated images, including lesion segmentation, biopsy results to specify benign and malignant cases, and BI-RADS categories to indicate the likelihood of malignancy. Besides, standardized partitions of training, validation, and test sets promote reproducibility and fair comparisons between different approaches. Thus, we present a publicly available BUS dataset whose novelty is the substantial increment of cases with the above-mentioned annotations and the inclusion of standardized partitions to objectively assess and compare CAD systems. ACQUISITION AND VALIDATION METHODS: The BUS dataset comprises 1875 anonymized images from 1064 female patients acquired via four ultrasound scanners during systematic studies at the National Institute of Cancer (Rio de Janeiro, Brazil). The dataset includes biopsy-proven tumors divided into 722 benign and 342 malignant cases. Besides, a senior ultrasonographer performed a BI-RADS assessment in categories 2 to 5. Additionally, the ultrasonographer manually outlined the breast lesions to obtain ground truth segmentations. Furthermore, 5- and 10-fold cross-validation partitions are provided to standardize the training and test sets to evaluate and reproduce CAD systems. Finally, to validate the utility of the BUS dataset, an evaluation framework is implemented to assess the performance of deep neural networks for segmenting and classifying breast lesions. DATA FORMAT AND USAGE NOTES: The BUS dataset is publicly available for academic and research purposes through an open-access repository under the name BUS-BRA: A Breast Ultrasound Dataset for Assessing CAD Systems. BUS images and reference segmentations are saved in Portable Network Graphic (PNG) format files, and the dataset information is stored in separate Comma-Separated Value (CSV) files. POTENTIAL APPLICATIONS: The BUS-BRA dataset can be used to develop and assess artificial intelligence-based lesion detection and segmentation methods, and the classification of BUS images into pathological classes and BI-RADS categories. Other potential applications include developing image processing methods like despeckle filtering and contrast enhancement methods to improve image quality and feature engineering for image description.

Magnetic resonance imaging-radioguided occult lesion localization (ROLL) in breast cancer using Tc-99m macro-aggregated albumin and distilled water control.

BACKGROUND: Magnetic resonance imaging (MRI) guided wire localization presents several challenges apart from the technical difficulties. An alternative to this conventional localization method using a wire is the radio-guided occult lesion localization (ROLL), more related to safe surgical margins and reductions in excision volume. The purpose of this study was to establish a safe and reliable magnetic resonance imaging-radioguided occult lesion localization (MRI-ROLL) technique and to report our initial experience with the localization of nonpalpable breast lesions only observed on MRI. METHODS: Sixteen women (mean age 53.2 years) with 17 occult breast lesions underwent radio-guided localization in a 1.5-T MR system using a grid-localizing system. All patients had a diagnostic MRI performed prior to the procedure. An intralesional injection of Technetium-99m macro-aggregated albumin followed by distilled water was performed. After the procedure, scintigraphy was obtained. Surgical resection was performed with the help of a gamma detector probe. The lesion histopathology and imaging concordance; the procedure's positive predictive value (PPV), duration time, complications, and accuracy; and the rate of exactly excised lesions evaluated with MRI six months after the surgery were assessed. RESULTS: One lesion in one patient had to be excluded because the radioactive substance came back after the injection, requiring a wire placement. Of the remaining cases, there were four malignant lesions, nine benign lesions, and three high-risk lesions. Surgical histopathology and imaging findings were considered concordant in all benign and high-risk cases. The PPV of MRI-ROLL was greater if the indication for the initial MR examination was active breast cancer. The median procedure duration time was 26 minutes, and all included procedures were defined as accurate. The exact and complete lesion removal was confirmed in all (100%) patients who underwent six-month postoperative MRI (50%). CONCLUSIONS: MRI-ROLL offers a precise, technically feasible, safe, and rapid means for performing preoperative MRI localizations in the breast.