الفهرس | Only 14 pages are availabe for public view |
Abstract Objective: This randomized controlled trial (RCT) aimed to detect the role of AI and DBT in assessment of mammography detected asymmetries Methods: This study was conducted at Women Imaging Unit, the department of radiology, faculty of medicine, Cairo University in the period from February 2021 to September 2021, the study included 52 patients who performed diagnostic or screening mammograms. All their mammograms revealed unilateral breast asymmetry.Their ages from 31 to 61 years with a mean age 43.96 ± 7.9 (mean ± SD). Results: Our results showed that DBT showed slightly higher specificity than mammography regarding breast asymmetry. DBT overcomes the problem of overlapping of the glandular tissue, so it can improve the diagnostic accuracy of mammography, adding Ultrasound to mammography improved the specificity of mammography and reduced its FP results as ultrasound has an established role in characterization of breast lesions. AI mammography software showed slightly lower sensitivity in our study, yet higher specificity compared to mammography, ultrasound, and Digital Breast Tomosynthesis. Conclusions: coupling AI to these modalities in the assessment of mammographic asymmetry may reduce the FP results, the unnecessary biopsies and increase the confidence of diagnosis. Also, AI may use as a decision assistance tool in determining the sono-mammographic diagnosis |