Search

This paper extends our previous method for COVID-19 diagnosis, proposing an enhanced solution for detecting COVID-19 from computed tomography (CT) images using a lean transfer learning-based model. To decrease model misclassifications, two key steps of image processing were employed. Firstly, the uppermost and lowermost slices were removed, preserving sixty percent of each patient's slices. Secondly, all slices underwent manual cropping to emphasize the lung areas. Subsequently, resized CT scans (224 × 224) were input into an Xception transfer learning model with a modified output. Both Xception's architecture and pretrained weights were leveraged in the method. A big and rigorously annotated database of CT images was used to verify the method. The number of patients/subjects in the dataset is more than 5000, and the number and shape of the slices in each CT scan varies greatly. Verification was made both on the validation partition and on the test partition of unseen images. Results on the COV19-CT database showcased not only improvement from our previous solution and the baseline but also comparable performance to the highest-achieving methods on the same dataset. Further validation studies could explore the scalability and adaptability of the developed methodologies across diverse healthcare settings and patient populations. Additionally, investigating the integration of advanced image processing techniques, such as automated region of interest detection and segmentation algorithms, could enhance the efficiency and accuracy of COVID-19 diagnosis., Competing Interests: No conflict of interest is to be reported., (Copyright © 2024 Kenan Morani et al.)

While the availability of "big data" on biophysical parameters through citizen science and/or from public/private sources is expected to help in addressing data scarcity issues, there is little understanding of whether and/or how such data will improve watershed simulations. This research aimed to evaluate whether improvements in resolutions of Digital Elevation Model (DEM) and soil data will enhance streamflow and sediment yield simulations and thereby improve soil and water management decisions. The study was conducted in two different-sized watersheds (Anjeni and Gilgel Abay with ~ 1 km 2 and ~ 1655 km 2 area, respectively) in the Upper Blue Nile basin in Ethiopia. Effects of DEM and soil data resolutions on streamflow and sediment yield were evaluated using the Soil and Water Assessment Tool (SWAT). The results showed that the effect of DEM and soil data resolution on streamflow and sediment yield simulation was scale dependent finer resolution DEM and soil datasets improved streamflow and sediment yield simulations in the smaller Anjeni watershed, whereas DEM resolution had no effect in the bigger Gilgel Abay watershed. Small watersheds are often used to understand watershed processes, and thus the use of finer-resolution spatial data for watershed simulations could result in better results. Findings from the smaller Anjeni watershed suggested that the combined use of finer resolution DEM and soil data could potentially improve sediment yield simulations although the lack of observed sediment yield data did not allow verification of this at the larger Gilgel Abay watershed., (© 2023. The Author(s), under exclusive licence to Springer Nature Switzerland AG.)

Water resource development opens up opportunities for improving smallholder farmer livelihoods in sub-Saharan Africa; however, implementation of water resource interventions to ensure sustainability hinges on the availability of sufficient quantity and quality data for monitoring, analysis and planning. Such data is often acquired through instrumentation of water resources (e.g. stream flow monitoring) or the use of hydrological models. In sub-Saharan Africa, data scarcity has limited the ability to monitor and make appropriate decisions for water resource allocation and use. Data derived from remote sensing has been considered a viable option to fill this gap; however, there is limited research in the region that evaluate the quality of the remotely sensed based datasets. This study evaluated actual evapotranspiration (AET) estimates derived from Advanced Very High Resolution Radiometer (AVHRR AET) images and Moderate Resolution Imaging Spectrometer (MOD16 AET) images using estimates from a grid-based Soil and Water Assessment Tool (SWAT). The SWAT model was set up for the entire country of Ethiopia, and calibrated and validated using observed streamflow at several meso-scale watersheds in which satisfactory model performance was obtained. AET estimates from the calibrated and validated SWAT model were then used to evaluate remotely sensed based AET for three landscapes. The AVHRR AET better agreed with the SWAT-simulated AET than the MOD16 AET, although the AVHRR AET overestimated the SWAT-simulated AET in all of the landscapes. Both remotely sensed AET products showed better agreement with the SWAT-simulated AET over agriculture dominated landscapes compared to grassland and forest dominated landscapes. The findings of the study suggest that remotely sensed based AET may help to fine-tune hydrological models in agricultural landscapes in data-scarce regions to improve studies on the impacts of water management interventions aiming to ensure environmental sustainability while enhancing agricultural production, and household income and nutrition., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020 The Authors. Published by Elsevier B.V. All rights reserved.)

The eastern Africa region has long been known for recurring drought, prolonged civil war and frequent pastoral conflicts. Several researchers have suggested that environmental factors can trigger conflicts among pastoralist communities, but quantitative support for this hypothesis is lacking. Here we use 29years of georeferenced precipitation and Normalized Difference Vegetation Index (NDVI) data to evaluate long term trends in scarcity of water and forage for livestock, and then ask whether these environmental stressors have any predictive power with respect to the location and timing of 11years of conflict data based on Armed Conflict Location and Event Data Project (ACLED) and Uppsala Conflict Data Program (UCDP). Results indicate that environmental stressors were only partly predictive of conflict events. To better understand the drivers behind conflict, the contribution of other potential stressors to conflict need to be systematically quantified and be taken into consideration., (Copyright © 2016 Elsevier B.V. All rights reserved.)

Modeling of suspended sediment emission into freshwater lakes is challenging due to data gaps in developing countries. Existing models simulate sediment concentration at a gauging station upstream and none of these studies had modeled total suspended solids (TSS) emissions by inflowing rivers to freshwater lakes as there are no TSS measurements at the river mouth in the upper Blue Nile basin. In this study a 10year TSS time series data generated from remotely sensed MODIS/Terra images using established empirical relationship is applied to calibrate and validate a hydrology model for Lake Tana in Upper Blue Nile Basin. The result showed that at a monthly time scale TSS at the river mouth can be replicated with Nash-Sutcliffe efficiency (NS) of 0.34 for calibration and 0.21 for validation periods. Percent bias (PBIAS) and ratio of the root-mean-square error to the standard deviation of measured data (RSR) are all within range. Given the inaccessibility and costliness to measure TSS at river mouths to a lake the results found here are considered useful for suspended sediment budget studies in water bodies of the basin., (Copyright © 2015 Elsevier B.V. All rights reserved.)