The computer-aided diagnosis (CAD) systems can highly improve the reliability and efficiency of melanoma recognition. As a crucial step of CAD, skin lesion segmentation has the unsatisfactory accuracy in existing methods due to large variability in lesion appearance and artifacts. In this work, we propose a framework employing multi-stage UNets (MS-UNet) in the auto-context scheme to segment skin lesion accurately end-to-end. We apply two approaches to boost the performance of MS-UNet. First, UNet is coupled with a context information fusion structure (CIFS) to integrate the low-level and context information in the multi-scale feature space. Second, to alleviate the gradient vanishing problem, we use deep supervision mechanism through supervising MS-UNet by minimizing a weighted Jaccard distance loss function. Four out of five commonly used performance metrics, including Jaccard index and Dice coefficient, show that our approach outperforms the state-ofthe-art deep learning based methods on the ISBI 2016 Skin Lesion Challenge dataset.
Myosin light chains (MLC) serve important regulatory functions in a wide range of cellular and physiological processes. Recent research found that MLC are also chromatin-associated nuclear proteins which regulate gene transcription. In this study, the MLC member myosin regulatory light chain 5 (MYL5) expression was upregulated in late stage cervical cancer patients, positively correlated with pelvic lymph node metastasis, and identified as a poor survival indicator. MYL5 overexpression promoted metastasis in cervical cancer in vitro and in vivo models, whereas MYL5 silencing had the converse effect. We demonstrated a bidirectional regulation between MYL5 and hypoxia inducible factor-1α (HIF-1α). HIF-1α activates MYL5 via binding to the hypoxia response element (HRE) in the promoter of MYL5, and MYL5 could sustain HIF-1α expression by tethering to recognition sequence AGCTCC in the HIF-1α promoter region. Clinical data confirmed a positive correlation between MYL5 and HIF-1α. In summary, our data show that MYL5 may act as a prognosis predictive factor in cervical carcinoma, and strategies that inhibit the interaction of MYL5 and HIF-1α may benefit the cervical carcinoma patients with metastasis.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.