Keywords:
Performed at one institution, Diagnostic or prognostic study, Retrospective, Cancer, Screening, Computer Applications-Detection, diagnosis, CAD, Mammography, Oncology, Breast, Artificial Intelligence, Artificial Intelligence and Machine Learning
Authors:
G. Porrello, A. Orlando, M. I. Schillaci, M. Dimarco, M. L. Di Vittorio, S. Busalacchi, M. insalaco, S. Vitabile, M. Midiri; Palermo/IT
DOI:
10.26044/ecr2020/C-05794
Purpose
Among females, breast cancer is the most commonly diagnosed cancer and the leading cause of cancer death. WHO data show that, in 2018 alone, over 2 million new cases of breast cancer were reported.[1]
To identify breast cancer at earlier stages, when treatment can be more successful, screening mammography programs were developed worldwide over the last decades. [2] A European multicentre study conducted by Tabár et al., demonstrated that in 29 years, mammographic screening provided a 30% reduction of breast cancer mortality [3].
While mammography has been proven to be a powerful tool against breast cancer, the accurate reading of mammograms is difficult and their interpretation is affected by high rates of false positives and false negatives. [2] Various data suggest that radiologists may miss up to 30% of breast cancers, especially in dense breasts, leading to a consistent number of interval cancers. [4]
With the aim to improve the efficacy of mammographic interpretation and diagnostic performance of breast cancer screening programs, computer-aided detection (CAD) softwares for mammography were developed and firstly approved by the Food and Drug Administration (FDA) in 1998. [5]
A CAD software is intended to assist and support radiologists as a "second reader" in identifying subtle lesions that might otherwise be missed. CAD marks potential areas of concern on the mammogram and radiologists will perform a visual check of those highlighted areas and, based on their own experience, decide if they need further work-up.
Recent studies on CAD systems for breast cancer and developing intelligent techniques have demonstrated an improvement in the detection rate of breast cancer, using CAD software, ranging from 4.7% to 19.5% when compared to radiologist performances without CAD. [6]
Our study aimed to assess the role of BD4BREAST, a Decisional Support System (DSS) software based on artificial intelligence with CAD functions, in identifying suspicious findings and assisting less experienced radiologists in the categorization of focal breast lesions according to the BI-RADS mammographic lexicon.