Breast imaging segmentation is a critical component of medical image analysis, playing a key role in breast cancer detection and diagnosis. An accurate and efficient segmentation model is essential for identifying and analyzing specific regions of interest within breast images. Through this study, we aim to provide state-of-the-art Segment Anything Model applications in breast imaging segmentation. The Segment Anything Model (SAM) was developed and presented by Meta AI in 2023 as the first foundational model for image segmentation. It was trained on natural image datasets and has shown good performance in accurately and efficiently identifying specific regions of interest. However, its effectiveness in segmenting medical images needs to be evaluated. After analyzing various studies in the literature regarding the use of SAM for segmenting breast imaging, we discovered that the performance of SAM varies considerably depending on the dataset and the task performed. For certain medical imaging datasets, its zero-shot segmentation can be moderate to poor, while for others it can be accurate. Tumor characteristics also have a significant impact on segmentation effectiveness. Larger and more contrasty tumors are more accurate because their edges are easier to see. On the other hand, complex shapes make segmentation harder, and the aspect ratio has minimal impact on this process.