• Clinical and Experimental Pediatrics
• /
• v.64 no.12
• /
• pp.632-641
• /
• 2021

Background: Over the past few decades, various goals have been defined to reduce the mortality of children caused by acute lower respiratory infections (ALRIs) worldwide. However, few spatial studies to date have reported on ALRI deaths. Purpose: We aimed to assess the spatial modeling of mortality from ALRI in children under 5 years of age during 2000-2017 using a global data. Methods: The data on the mortality of children under 5 years old caused by ALRI were initially obtained from the official website of the World Health Organization. The income status of their home countries was also gathered from the Country Income Groups (World Bank Classification) website and divided into 5 categories. After that, in the ArcGIS 10.6 environment, a database was created and the statistical tests and related maps were extracted. The Global Moran's I statistic, Getis-Ord Gi statistic, and geographically weighted regression were used for the analyses. In this study, higher z scores indicated the hot spots, while lower z scores indicated the cold spots. Results: In 2000-2017, child mortality showed a downward trend from 17.6 per 100,000 children to 8.1 and had a clustered pattern. Hot spots were concentrated in Asia in 2000 but shifted toward African countries by 2017. A cold spot that formed in Europe in 2007 showed an ascending trend by 2017. Based on the results of geographically weighted regression test, the regions identified as the hot spots of mortality from ALRI in children under 5 years old were among the middle-income countries (R²=0.01, adjusted R²=8.77). Conclusion: While the total number of child deaths in 2000-2017 has decreased, the number of hot spots has increased among countries. This study also concluded that, during the study period, Central and Western Africa countries became the main new hot spots of deaths from ALRI.

• Geophysics and Geophysical Exploration
• /
• v.26 no.2
• /
• pp.62-76
• /
• 2023

Recently, machine learning (ML) techniques have been actively applied for seismic trace interpolation. However, because most research is based on training-inference strategies that treat missing trace gather data as a 2D image with a blank area, a sufficient number of fully sampled data are required for training. This study proposes trace interpolation using ML, which uses only irregularly sampled field data, both in training and inference, by modifying the training-inference strategies of trace-based interpolation techniques. In this study, we describe a method for constructing networks that vary depending on the maximum number of consecutive gaps in seismic field data and the training method. To verify the applicability of the proposed method to field data, we applied our method to time-migrated seismic data acquired from the Vincent oilfield in the Exmouth Sub-basin area of Western Australia and compared the results with those of the conventional trace interpolation method. Both methods showed high interpolation performance, as confirmed by quantitative indicators, and the interpolation performance was uniformly good at all frequencies.