BCGAN: A CGAN-based over-sampling model using the boundary class for data balancing

The Journal of Supercomputing - A class imbalance problem occurs when a dataset is decomposed into one majority class and one minority class. This problem is critical in the machine learning...     

A computer-aided diagnostic system for liver tumor detection using modified U-Net architecture

The Journal of Supercomputing - A multimedia-based medical decision-making system is an ultimate requirement in the medical imaging domain. In the healthcare sector, achieving quick and efficient...     

Modeling investment priorities for national road improvements: A case study of Korea

The paper addresses specific problems associated with formulating investment priorities for national road improvements in Korea. More specifically, the objectives of the paper are (1) to present a model for formulating investment priorities for road improvements that may be useful to developing countries, (2) to test the model with simplified data for deriving investment priorities for Korea's road improvements, and (3) to learn lessons that can be applicable to other developing countries with particular respect to recommending investment priorities for road improvements. The study results are useful for planners, in developing countries, and also for evaluation analysts in funding institutions who are confronted with evaluating investment priorities for deteriorating road conditions.     

Whole Exome Sequencing for a Patient with Rubinstein-Taybi Syndrome Reveals de Novo Variants besides an Overt CREBBP Mutation

Rubinstein-Taybi syndrome (RSTS) is a rare condition with a prevalence of 1 in 125,000–720,000 births and characterized by clinical features that include facial, dental, and limb dysmorphology and growth retardation. Most cases of RSTS occur sporadically and are caused by de novo mutations. Cytogenetic or molecular abnormalities are detected in only 55% of RSTS cases. Previous genetic studies have yielded inconsistent results due to the variety of methods used for genetic analysis. The purpose of this study was to use whole exome sequencing (WES) to evaluate the genetic causes of RSTS in a young girl presenting with an Autism phenotype. We used the Autism diagnostic observation schedule (ADOS) and Autism diagnostic interview revised (ADI-R) to confirm her diagnosis of Autism. In addition, various questionnaires were used to evaluate other psychiatric features. We used WES to analyze the DNA sequences of the patient and her parents and to search for de novo variants. The patient showed all the typical features of Autism, WES revealed a de novo frameshift mutation in CREBBP and de novo sequence variants in TNC and IGFALS genes. Mutations in the CREBBP gene have been extensively reported in RSTS patients, while potential missense mutations in TNC and IGFALS genes have not previously been associated with RSTS. The TNC and IGFALS genes are involved in central nervous system development and growth. It is possible for patients with RSTS to have additional de novo variants that could account for previously unexplained phenotypes.     

Stimuli-Responsive Dynamic Metaholographic Displays with Designer Liquid Crystal Modulators

Flat optics, realized by the artificially created 2D material platform called optical metasurfaces, is currently undergoing a science-to-technology transition. However, “real-time” active operations of such flat optical devices remain yet unresolved. Here, liquid crystals (LCs)-integrated metaholograms for ultracompact dynamic holographic displays are proposed. The anisotropic nature of the LCs allows facile and repeatable manipulation of the polarization of light. Specifically designed (“designer”) LCs and efficient helicity-encoded metaholograms are combined to realize stimuli-responsive dynamic displays. The designer LC modulators are used as switches that enable a variety of external stimuli (e.g., electric field, heat, surface pressure) to operate holographic images in real-time. Such a dynamic metaholographic platform will provide a path to external stimuli-driven “smart” sensing and display applications such as hologram labels for temperature/pressure/touch monitoring and interactive holographic displays with haptic motion recognition.     

Monitoring skin condition using life activities on the SNS user documents

Multimedia Tools and Applications - Social networks are not new to the IT landscape. Starting from bulletin boards and chat rooms, they have evolved to include desktop and mobile device...