• Anna Carolina Falanga Brenha
• Pedro Cunha Carneiro
• Ana Claudia Patrocínio

In a scenario where breast cancer is still a worldwide challenge to healthcare professionals, with more than half a million deaths registered annually, the early detection of the disease faces a range of obstacles. The need of great image contrast quality, often associated with high radiation doses, is opposed by the constant efforts to reduce radiation absorbance for patients and technicians during breast imaging exams, as outlined in ALARA (As Low As Reasonably Achievable) guidelines. The present study proposed a Pix-2-Pix neural network implementation as an image contrast enhancement tool, projected to reproduce high-dose contrast quality in low-dose images, implementing the necessary conditions for proper diagnosis while preserving ALARA standards for patient and technician safety. The Carneiro Contrast Index (CCI) was employed for quantitative assessments, demonstrating significant contrast improvement without additional radiation. Also, the results successfully validate this technique’s potential to support safer breast cancer screening clinical practices. Further works include different datasets validation and optimization of other learning parameters.

2D Mammogram, Artificial Intelligence Breast Cancer, Image Contrast, Neural Networks

"Pix-2-Pix Neural Network Proposal for Automatic Breast Images Contrast Correction", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 1, page no.e699-e705, January-2025, Available :http://www.jetir.org/papers/JETIR2501583.pdf

"Pix-2-Pix Neural Network Proposal for Automatic Breast Images Contrast Correction", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 1, page no. ppe699-e705, January-2025, Available at : http://www.jetir.org/papers/JETIR2501583.pdf