An overview of preprocessing techniques for mammography breast cancer prediction
Vishwayogita A. Savalkar, Gurpreet Singh Saini, Shivaji D. Pawar · 2025
Breast cancer is a deadly disease that usually affects women. Improving the prognosis of patients with breast cancer requires early and precise diagnosis. Although mammography is still the gold standard for screening, picture noise, and breast density can make it difficult to interpret mammograms. Preprocessing methods are essential for optimizing the performance of breast cancer prediction models and preparing mammograms for further study. Enhancing the quality of mammography data is crucial for improving the prediction power of breast cancer models. The several preprocessing techniques utilized in the analysis of mammograms to predict breast cancer are thoroughly evaluated and analyzed in this paper. We classify these approaches into image registration, noise reduction, contrast enhancement, segmentation, and pectoral muscle removal techniques and provide a comparative analysis of their benefits, drawbacks, and applicability in clinical settings. Overall, this survey used several published articles to assess the current capability of preprocessing and segmentation algorithms, with accuracy for Breast Cancer Prediction as a result provided.