We propose a novel unsupervised domain adaptation framework based on domain-specific batch normalization in deep neural networks. We aim to adapt to both domains by specializing batch normalization layers in convolutional neural networks while allowing them to share all other model parameters, which is realized by a twostage algorithm. In the first stage, we estimate pseudolabels for the examples in the target domain using an external unsupervised domain adaptation algorithm-for example, MSTN [27] or CPUA [14]-integrating the proposed domain-specific batch normalization. The second stage learns the final models using a multi-task classification loss for the source and target domains. Note that the two domains have separate batch normalization layers in both stages. Our framework can be easily incorporated into the domain adaptation techniques based on deep neural networks with batch normalization layers. We also present that our approach can be extended to the problem with multiple source domains. The proposed algorithm is evaluated on multiple benchmark datasets and achieves the state-of-theart accuracy in the standard setting and the multi-source domain adaption scenario.
Non-alcoholic fatty liver disease (NAFLD) affects a substantial proportion of the world population, and its prevalence has been increasing. The study was aimed at evaluating the prevalence and peri-transplant risk factors for post-liver transplantation (LT) NAFLD. A retrospective review was performed for adult recipients who underwent late protocol biopsy (>1 yr after LT) between August 2010 and December 2012. Hepatic steatosis was reviewed and graded by hepatopathologists, and the peri-transplant factors were analyzed for relationships to histologically proven NAFLD. Total 166 biopsies had been performed in 156 recipients. NAFLD was present in 27.1% at a mean period of 35.4 months between LT and biopsy, moderate and severe steatosis (≥33%) consisted of 28.9%. In multivariate analysis, pre-LT alcoholic cirrhosis (odds ratio [OR] 8.031, p = 0.003), obesity at biopsy (OR 3.873, p = 0.001), and preexisting donor graft steatosis (OR 3.147, p = 0.022) were significant risk factors for post-LT NAFLD. In conclusion, NAFLD represented a considerable portion of recipients, but this prevalence was not higher than those for general population. Three risk factors were significantly related to post-LT NAFLD, and recipients with those factors should be monitored for NAFLD. Furthermore, possible progression to non-alcoholic steatohepatitis (NASH) or fibrosis and metabolic syndrome should be considered in future studies.
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.