BREAST DENSITY EVALUATION: A COMPARISON BETWEEN ASSESSMENT BY A RADIOLOGIST AND THE COMPUTER-ASSISTED THRESHOLD TECHNIQUE

Traditionally, mammographic density (MD) of the breast has been assessed by a radiologist visually. This subjective evaluation requires significant experience to distinguish the relative proportions of the fibrous connective tissue and adipose tissue in the mammary gland correctly. The aim of this study is to compare the capabilities of the different methods (visual and computer-assisted) for assessing mammographic density. Our sample in this study consists of 66 patients with digital mammography. The mammographic density has been evaluated using the four-grade scale of the American College of Radiology (ACR); visually, visually using an analog scale and semi-automated using UTHSCSA Image Tool 3.0, Image J and Adobe Photoshop CS6 software. The average mammographic density calculated using the different methods is as follows: 34.8% (from 10% to 70%); 32.1% (from 10% to 60%); 23% (from 0% to 70.9%); 22.7% (from 2.5% to 78.1%) and 22.5% (from 1.5% to 72.4%). There is a strong correlation between the results obtained visually and those calculated using a computer-assisted measurement (p< 0.0001). A strong correlation was found also between the results acquired using the different semi-automated programs (p< 0.0001). Precise measurement of mammographic density is of great importance for the mammographic screening and evaluation of breast cancer risk. The semi-automated methods, used for this purpose are objective, accessible and reproducible tools and have some advantages over the subjective visual assessment.