Increasing dietary diversity may be an approach to reduce the burden of stunting and chronic malnutrition among young children in regions with seasonal food insecurity.
Carbapenem-resistantEnterobacteriaceaeconferred by New Delhi metallo-b-lactamase (NDM-1) resistance mechanism are endemic in India and Southeast Asia. An understanding of risk factors for NDM-1 infections is necessary to guide prevention strategies. We performed a retrospective case-control study of patients admitted at Christian Medical College Hospital, Vellore, India between May 2010 and August 2014 withKlebsiella pneumoniaeblood stream infection (BSI). We compared patients with BSI caused by NDM-1 producing strains to two control groups: BSI with other multidrug resistant (MDR) strains and BSI with pan-susceptible strains. The study groups were assessed for risk factors for the outcomes: (1) infection with any MDR strain compared to pan-susceptible; and, (2) infection with NDM-1 strain as compared with other MDR and (3) Mortality. A total of 101 patients with BSI with NDM-1 producingKlebsiella pneumoniaewere matched to two groups of controls: 112 with non-NDM-1 MDR strains and 101 with pan-susceptible strains. Medical (OR 10.4) and neonatal (OR 0.7) ICU admission, central venous catheter placement (CVC, OR 7.4) predicted MDR BSI. Prior carbapenem use (OR 8.4) and CVC (OR 4.8) predicted acquisition of an NDM-1 strain. Significant predictors for mortality included ICU stay (OR 3.0), mechanical ventilation (OR 3.2), female gender (OR 2.2), diabetes (OR 0.4). CVC placement, prior carbapenem use and ICU admission were significantly associated with BSI with NDM-1 producing and other MDR strains.
In this work, we propose a special cascade network for image segmentation, which is based on the U-Net networks as building blocks and the idea of the iterative refinement. The model was mainly applied to achieve higher recognition quality for the task of finding borders of the optic disc and cup, which are relevant to the presence of glaucoma. Compared to a single U-Net and the state-of-the-art methods for the investigated tasks, the presented method outperforms others by multiple benchmarks without a need for increasing the volume of datasets. Our experiments include comparison with the best-known methods on publicly available databases DRIONS-DB, RIM-ONE v.3, DRISHTI-GS, and evaluation on a private data set collected in collaboration with University of California San Francisco Medical School. The analysis of the architecture details is presented. It is argued that the model can be employed for a broad scope of image segmentation problems of similar nature.
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.