Tehran University of Medical Sciences Frontiers in Biomedical Technologies 2345-5837 12 4 2025 10 01 812 829 Kimia Jalalian Department of Biomedical Engineering, Pooyesh Institute of Higher Education, Qom Golnaz Hosseini Department of Biomedical Engineering, Pooyesh Institute of Higher Education, Qom, Iran Razieh Ghiasi Department of Computer Engineering, Qom University, Qom, Iran Alireza Bosaghzadeh Department of Computer Engineering, Shahid Rajaee Teacher Training University Tehran, Iran 2024 06 19 2025 05 08 Purpose: Manually segmenting mammograms is time-consuming and subjective. Hence, developing an automatic method to address challenges like low signal-to-noise ratio, various mass shapes and sizes, and high false positive rates is crucial. In this study, we present an automated approach for mass segmentation to address these challenges effectively. Materials and Methods: Our proposed system consists of two phases: the pre-processing phase, which includes denoising, contrast enhancement, image cropping, resizing, and augmentation of mammograms; and the model design phase, where UNet++ is employed as an encoder-decoder-based network for segmenting breast masses. The encoder captures relevant information from various regions in the input image, while the decoder reconstructs the spatial location of the target region. We extensively experimented with a publicly accessible CBIS-DDSM dataset to evaluate our proposed system performance. Results: Based on our findings, our proposed method demonstrates promising results with a precision rate of 91.84%, a True Positive Rate of rate of 93.66%, a Dice Score Coefficient measuring 92.66%, and a Jaccard Index of 86.46%. Conclusion: The use of UNet++ combined with a pre-processing pipeline in digital mammography has shown promising results in accurately segmenting breast masses and has the potential to significantly improve early breast cancer detection. https://fbt.tums.ac.ir/index.php/fbt/article/view/1045 https://fbt.tums.ac.ir/index.php/fbt/article/download/1045/515