High-resolution synthesis of high-density breast mammograms: Application to improved fairness in deep learning based mass detection.

Computer-aided detection systems based on deep learning have shown good performance in breast cancer detection. However, high-density breasts show poorer detection performance since dense tissues can mask or even simulate masses. Therefore, the sensitivity of mammography for breast cancer detection can be reduced by more than 20% in dense breasts. Additionally, extremely dense cases reported an increased risk of cancer compared to low-density breasts. This study aims to improve the mass detection performance in high-density breasts using synthetic high-density full-field digital mammograms (FFDM) as data augmentation during breast mass detection model training. To this end, a total of five cycle-consistent GAN (CycleGAN) models using three FFDM datasets were trained for low-to-high-density image translation in high-resolution mammograms. The training images were split by breast density BI-RADS categories, being BI-RADS A almost entirely fatty and BI-RADS D extremely dense breasts. Our results showed that the proposed data augmentation technique improved the sensitivity and precision of mass detection in models trained with small datasets and improved the domain generalization of the models trained with large databases. In addition, the clinical realism of the synthetic images was evaluated in a reader study involving two expert radiologists and one surgical oncologist.

Advantages of preoperative localization and surgical resection of metastatic axillary lymph nodes using magnetic seeds after neoadjuvant chemotherapy in breast cancer.

PURPOSE: To assess the safety and effectiveness of magnetic seeds in preoperative localization and surgical dissection of metastatic axillary lymph nodes (LN+) in breast cancer patients with axillary involvement, after neoadjuvant chemotherapy (NAC). In addition, to assess the impact of targeted axillary dissection (TAD) in reducing the rate of false negatives (FN) in sentinel lymph node biopsy (SLNB). MATERIALS AND METHODS: A cross-sectional prospective cohort study was conducted from April 2017 to September 2019, including breast cancer patients with axillary lymph node involvement treated with NAC. Prior to NAC, the LN+ were marked by ultrasound-guided clip insertion. After NAC, a magnetic seed (Magseed®) was inserted in the clip-marked lymph node (MLN). During surgery, the MLN was located and removed with the aid of a magnetic detection probe (Sentimag®) and the sentinel lymph node was removed. Axillary lymph node dissection (ALND) was used to determine the rate of FN for SLNB alone and the combination of SLNB and MLN dissection, called TAD. RESULTS: The study included 29 patients (mean age, 55; range, 30-78 years). Selective preoperative localization and surgical dissection were successful for all 30 MLNs (100%). The MLN corresponded to the SLN in 50% of cases. After ALND, there were 21.4% (3/14) FN with SLNB alone and 5.9% (1/17) with TAD. CONCLUSIONS: Following NAC, selective surgical removal of MLN by preoperative localization using magnetic seeds is a safe and effective procedure with a success rate of 100%. Adding TAD reduces the rate of FN associated with SLNB alone.