Enhancing mammographic imaging

a novel multiscale morphological approach

Resumen

Accurate interpretation of mammographic images is crucial for the early detection and diagnosis of breast cancer. Unfortunately, the quality of these images is often compromised by noise, low resolution, and insufficient contrast, hindering the identification of key features such as microcalcifications and increasing the risk of misdiagnosis or the need for repeat examinations. This work introduces an innovative approach for the enhancement of mammographic images through the Multiscale Top-Hat Transform by Reconstruction (MTHR), a multiscale morphological method that significantly improves clarity, contrast, and visibility of important structures while maintaining the integrity of essential information. Unlike existing techniques, MTHR extracts and maximizes multiple features, using top-hat transform by reconstruction to add bright areas and subtract dark zones to enhance contrast and fine details. The results obtained surpass several state-of-the-art contrast enhancement algorithms, demonstrating significant improvements in clarity and visibility of key structures.