Abstract
PURPOSE: Inflammatory breast cancer (IBC) is a rare and aggressive type of breast cancer, as many physicians may not be aware of it in terms of symptoms and diagnosis. Mammography is the first choice in breast screenings and diagnosis. Because of a lack of expertise and imaging datasets, IBC portrayal and machine learning-based diagnosis systems have not yet been studied thoroughly. Developing scanning and diagnosis tools can close the knowledge gap and barriers to a timely IBC diagnosis. MATERIALS AND METHODS: The dataset includes 20 women aged 34-75, of whom 10 were clinically diagnosed with IBC and 10 with non-IBC. A breast mapping and scanning model was developed. Gray-level co-occurrence matrices were used to characterize skin thickening, edema, breast density, microcalcifications, and breast size asymmetry in bilateral mammographic images. RESULTS: A one-way analysis of variance (ANOVA) test was performed to evaluate differences between mammogram breasts with IBC, non-IBC, and healthy breasts. Higher breast density variations were calculated in breasts with IBC in the anterior (P = 0.0147) and middle (P = 0.0026) regions. Breasts with IBC showed higher microcalcifications (P = 0.0472) than the other breasts, and bilateral analyses showed higher variations (P = 0.1367). Breast size asymmetry (P = 0.9833) was not significantly different between the groups. CONCLUSION: Skin thickening, edema, and breast density-related parameters were found to be associated with IBC. This study thus lays the foundation of machine learning diagnosis models for IBC.