Characterization of Breast Fibroadenomaand Cyst by UsingUltrasonographyand TextureAnalysis

Abstract

Breast Fibroadenoma is a hormone-dependent neoplasm of early reproductive life that lactates during pregnancy and involutes along with the rest of the breast in per menopause. The peak incidence is between the ages of 15 and 35 years . Image texture is a function of the spatial variation of pixel intensities in an image .Image texture analysis can provide quantitative information in the form of texture features that is not visible to human vision.The general objective of this study was to characterize breast Fibroadenoma in U/S image by using texture analysis in order differentiate between the different tissue morphology of breast lesions. This study was done to determine the ultrasonography characteristics in breast (Fibroadenoma ,cyst ) versus normal fibroglandular breast tissue . It is carried out in Saad Specialist hospital K.S.A , in patient with breast lesions that referred to radiology department in the period from January 2014 to May 2016. A total of 130 patients were included in this study. US result showed, 41% with regular outline indicate the benignancy of disease, 30.7% with normal result and more than 95% with breast feeding, 46.5% form all women having hypoechoic lesion, with homogenous echotexture in 54.4% of the patient, no calcification seen, 65% avascular and prevascularity was noted in 0.9% of lesion mostly near the vascularity of breast. UOQ commonly affected in both breast, Linear discriminant analysis was used for the tissue classification. The study found that the Fibroadenoma texture reveal a different underlying pattern compared to the cyst and normal breast tissues with classification sensitivity and specificity 100% and 92.5% respectively for Fibroadenoma, and the combination of the texture features throughout the different U/S image phases provide the highest predictive overall accuracy of 95% using linear discriminant analysis technique.