Impact of Tomosynthesis Acquisition on 3D Segmentations of Breast Outline and Adipose/Dense Tissue with AI: A Simulation-Based Study.

Digital breast tomosynthesis (DBT) reconstructions introduce out-of-plane artifacts and false-tissue boundaries impacting the dense/adipose and breast outline (convex hull) segmentations. A virtual clinical trial method was proposed to segment both the breast tissues and the breast outline in DBT reconstructions. The DBT images of a representative population were simulated using three acquisition geometries: a left-right scan (conventional, I), a two-directional scan in the shape of a "T" (II), and an extra-wide range (XWR, III) left-right scan at a six-times higher dose than I. The nnU-Net was modified including two losses for segmentation: (1) tissues and (2) breast outline. The impact of loss (1) and the combination of loss (1) and (2) was evaluated using models trained with data simulating geometry I. The impact of the geometry was evaluated using the combined loss (1&2). The loss (1&2) improved the convex hull estimates, resolving 22.2% of the false classification of air voxels. Geometry II was superior to I and III, resolving 99.1% and 96.8% of the false classification of air voxels. Geometry III (Dice = (0.98, 0.94)) was superior to I (0.92, 0.78) and II (0.93, 0.74) for the tissue segmentation (adipose, dense, respectively). Thus, the loss (1&2) provided better segmentation, and geometries T and XWR improved the dense/adipose and breast outline segmentations relative to the conventional scan.

An interval prototype classifier based on a parameterized distance applied to breast thermographic images.

Breast cancer is one of the leading causes of death in women. Because of this, thermographic images have received a refocus for diagnosing this cancer type. This work proposes an innovative approach to classify breast abnormalities (malignant, benignant and cyst), employing interval temperature data in order to detect breast cancer. The learning step takes into account the internal variation of the intervals when describing breast abnormalities and uses a way to map these intervals into a space where they can be more easily separated. The method builds class prototypes, and the allocation step is based on a parameterized Mahalanobis distance for interval-valued data. The proposed classifier is applied to a breast thermography dataset from Brazil with 50 patients. We investigate two different scenarios for parameter configuration. The first scenario focuses on the overall misclassification rate and achieves 16 % misclassification rate and 93 % sensitivity to the malignant class. The second scenario maximizes the sensitivity to the malignant class, achieving 100 % sensitivity to this specific class, along with 20 % overall misclassification rate. We compare the performances of our approach and of many methods taken from the literature of interval data classification for the breast thermography task. Results show that our method outperforms competing algorithms.