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

Enhancing the quality of digital breast images is crucial for improving the visualization and identification of breast lesions. Digital image processing techniques play a fundamental role in contrast enhancement, which aids in distinguishing fibroglandular and cancerous tissues, reducing diagnostic subjectivity. Subtle indicators of breast cancer, such as small masses and microcalcifications, can be challenging to detect in low-dose 2D mammograms, where noise further complicates lesion differentiation. Studies highlight the importance of integrating noise filtering and contrast enhancement methods to improve image clarity. The effectiveness of contrast enhancement is often evaluated through qualitative visual inspection, which remains subjective. Consequently, various quantitative contrast performance measures, including variance, MSSIM (Mean Structural Similarity Index Measure), and PSNR (Peak Signal-to-Noise Ratio), have been proposed, though they do not always align with visual perception. Signal-to-Noise Ratio (SNR) and Enhancement Measure Evaluation (EME) are widely used contrast quantification metrics, yet they present limitations when applied to mammographic images due to their dependence on specific object characteristics or predefined regions of interest. To address the need for a global metric capable of objectively assessing contrast enhancement across an entire image, this study introduces the Carneiro Contrast Index (CCI). CCI computes local contrast through a 3x3 kernel, generating a contrast map that highlights regions of high and low contrast. The arithmetic mean of standard deviation values from this matrix provides a single global contrast index. Unlike EME, which requires region-of-interest selection, CCI enables whole-image analysis, ensuring greater reproducibility and applicability across different imaging conditions. The proposed metric aligns with radiologists' visual assessments and established contrast measures, demonstrating its robustness in evaluating contrast enhancement techniques. Future research will explore the integration of CCI with deep learning models and automated diagnostic systems, further enhancing breast cancer detection accuracy and efficiency.

2D Mammogram, Contrast Evaluation, Image Processing, Medical Imaging

"Objective Contrast Evaluation in Mammograms: The Carneiro Contrast Index Approach", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 2, page no.c726-c733, February-2025, Available :http://www.jetir.org/papers/JETIR2502284.pdf

"Objective Contrast Evaluation in Mammograms: The Carneiro Contrast Index Approach", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 2, page no. ppc726-c733, February-2025, Available at : http://www.jetir.org/papers/JETIR2502284.pdf