Ultrasound Diagnosis of Breast Cancer

Application of a Novel Microvascular Imaging Technique in Breast Lesion Evaluation

“…Several studies have found that flow distribution patterns in malignant lesions differ from those in benign lesions. Peripheral, linear and regular flow signals appear to exist in benign lesions; whereas central distribution, branching and irregular flow signals strongly suggest malignant lesions (Blohmer et al 1999;Gokalp et al 2009;Milz et al 2001;Sencha et al 2013;Sirous et al 2015;Stanzani et al 2014). Both location and morphology of the microvessels were accounted for in our study.…”

“…Conventional CDI and PDI have facilitated the diagnosis of breast lesions for more than 20 y. Consistently, these flow techniques detect more flow signals in malignant breast lesions (Blohmer et al 1999;Davoudi et al 2014;Milz et al 2001;Stanzani et al 2014), especially in the center, with more branching and irregular vessels (Milz et al 2001;Sencha et al 2013;Stanzani et al 2014). Newly formed vessels penetrating the lesion from the periphery to the center can be readily observed.…”

Benefits of Superb Microvascular Imaging in Vascularity Evaluation of Breast Masses: Diagnostic Performance Comparision with Color and Power Doppler Imaging

“…Bu bulgular, daha önceki çalışmalarla benzerlik gösteriyor. 24,25 Bu çalışma, vasküler dağılımı tanımlama, internal vaskülarizasyonu gösterme ve internal vaskülarite özelliklerini değerlendirmede rSMG'nin daha etkili ve doğru olduğunu göstermiştir. Sonuç rSMG; RDG ve PDG'den daha fazla vasküler dallanma detayı göstermekte olup benign-malign meme lezyonunun ayırt edilmesinde uygulanabilir bir yöntemdir.…”

A generic deep learning framework to classify thyroid and breast lesions in ultrasound images

“…A possible association between breast and thyroid cancer has also been demonstrated, including shared hormonal risk factors and genetic susceptibility [15]. Furthermore, thyroid and breast cancers do share common image characteristics under high frequency ultrasound scans such as malignant lesions with a taller-than-wide shape ratio, hypo-echogenicity, and ill-defined margins [16,17]. This observation provides a strong motivation for developing a generic convolutional neural network (CNN) model that can be used to classify breast and thyroid cancers.…”