Abstract—Breast cancer is the leading type of cancer among women worldwide, but it can be cured if diagnosed at an early stage. Mammography is the main means of screening for cancer and provides useful information on the signs of cancer, such as macrocalcifications, masses, architectural distortion and asymmetry. This information is not easy to distinguish because of certain defects in mammographic images, including low contrast, high noise, blurring and confusion. To solve this problem, a new region-of-interest detection strategy based on multi-scale product and segmentation has been developed. The aim of this strategy is to improve the visibility of microcalcifications in regions of interest for better detection and interpretation. Mammographic images are represented in a way that facilitates the detection of microcalcifications and the classification of healthy and cancerous images. Using Matlab software, the algorithms developed were used on images from the public accessible international database of the Mammographic Image Analysis Society (MIAS), which contains 322 images. The results obtained in terms of visual quality were satisfactory. Compared with the results of the literature, our results appear to be better. 

Index Terms—breast cancer, mammography, multi-scale product, region of interest, segmentation

Cite: Nadji A. Yeltongar, Jérôme B. Mbainaibeye, and Toussaint C. Gnonwa, "Detection of the Regions of Interest in Mammographic Images by the Multiscale Product and Segmentation Using the Continuous Wavelet Transform," International Journal of Signal Processing Systems, Vol. 12, pp. 7-11, 2024. doi: 10.18178/ijsps.12.7-11

Copyright @ 2024  by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.