Mammographic density classification applying the CLAHE contrast enhancement method

Breast cancer is one of the top causes of death among women. To assist radiologists in making an early diagnosis, systems for computer-aided diagnosis (CAD) are being developed. We offer a diagnostic framework based on ResNet-50, a widely used clinical breast screening tool that is based on mammographic imaging, for the precise classification of breast abnormalities. The craniocaudal (CC) and mediolateral oblique (MLO) views on screening mammography provide two independent perspectives of each breast. We employ two views for breast classification since they are complimentary and dual-view-based approaches are effective. A deep model is trained using a loss function, and we utilize the focused loss function since it concentrates on learning challenging cases. Breast cancer affects over 1.5 million women worldwide each year. In women, breast cancer is the most prevalent malignant tumor. The highest intensive rate of the disease (per 100,000 women) is 30–40% in England, Denmark, the USA, and the Baltic countries, and the lowest rate is 2–10% in countries such as Japan and Mexico. This disease rarely occurs in women under the age of 25–30, but as the age limit increases, the number of diseases increases, and it reaches its “peak” at the age of 50–60. In recent years, the number of cases of this type of tumor disease in Uzbekistan has gradually increased, and it is now the leading cause of malignant tumors in women [1,2]. The World Health Organization's most recent statistics show that the incidence of breast cancer was the highest among oncological diseases in Uzbekistan in 2018 with a death rate of 1,449, or 0.92% of all deaths [2]. In 1993, there were 5.3 cases of this illness per 100,000 persons, but by 1998, that number had risen to 6.1. Every year, more people with breast cancer are receiving a diagnosis. 154 people with this condition were listed in the primary registry in the Samarkand region alone in 2000, and 300 patients were listed there in 2019.

Перевод пока недоступен