Breast ultrasound (BUS) image segmentation is challenging and critical for BUS computer-aided diagnosis (CAD) systems. Many BUS segmentation approaches have been studied in the last two decades, but the performances of most approaches have been assessed using relatively small private datasets with different quantitative metrics, which results in a discrepancy in performance comparison. Therefore, there is a pressing need for building a benchmark to compare existing methods using a public dataset objectively, to determine the performance of the best breast tumor segmentation algorithm available today, and to investigate what segmentation strategies are valuable in clinical practice and theoretical study. In this work, a benchmark for B-mode breast ultrasound image segmentation is presented. In the benchmark, (1) we collected 562 breast ultrasound images and proposed standardized procedures to obtain accurate annotations using four radiologists; (2) we extensively compared the performance of 16 state-of-the-art segmentation methods and demonstrated that most deep learning-based approaches achieved high dice similarity coefficient values (DSC ≥ 0.90) and outperformed conventional approaches; (3) we proposed the losses-based approach to evaluate the sensitivity of semi-automatic segmentation to user interactions; and (4) the successful segmentation strategies and possible future improvements were discussed in details.
Background: Breast cancer (BC) is one of the most prevalent malignancies among women globally. Emerging evidence indicates that long non-coding RNAs (lncRNAs) are associated with BC carcinogenesis. In the current study, we explored the mechanism by which LINC00662 regulates BC. Methods: Quantitative real-time PCR (qRT-PCR) assessed RNA expressions while western blot for protein levels. Kaplan Meier analysis evaluated overall survival (OS). Cytoplasmic/nuclear fractionation, RNA binding protein immunoprecipitation (RIP) and luciferase reporter assays probed into the underlying molecular mechanism of LINC00662 in BC. Xenograft model was established to explore the influence of LINC00662 on BC progression in vivo. R square graphs were utilized to represent RNA relationships. Results: LINC00662 is overtly overexpressed in BC tissues and cell lines. LINC00662 knockdown hampers cell proliferation, migration, invasion and stemness. LINC00662 expression is negatively correlated with OS of BC patients. LINC00662 up-regulates SOX2 expression by competitively binding to miR-144-3p, thereby modulating BC cell progression. Xenograft experiments verified that LINC00662 promotes BC tumor growth and cell stemness in vivo. Conclusion: LINC00662 enhances cell proliferation, migration, invasion and stemness in BC by targeting miR-144-3p/SOX2 axis. The findings in the present study suggested that LINC00662 could be a potential therapeutic target for BC treatment. Graphical
10 March 2025 in Journal of Imaging Informatics in Medicine