Search

Glaciers are under pressure in the current climate warming trend. The aerial extent and the mass balance, a measure of mass loss or mass gain of a glacier are representative of the “health state” of a glacier. They are recognized as Essential Climate Variables by the World Meteorological Organization. Due to the sheer number of glaciers of roughly 200.000 worldwide and their inaccessibility, satellite data has been proven as a valuable source to map glaciers around the globe. Satellite imagery has enabled researchers to map the extent of glaciers and glacier surface types (glacier facies, i.e. snow, firn, ice). Multi-temporal information on the glacier surface elevation and mass balance proxies (snow cover, albedo) which are highly correlated with the glacier mass balance have been used to estimate mass changes. Today the ever-growing amount of satellite data motivates the development of new methodology to maximize the information retrieved (value-added) and ease the handling and storage of the data in data cubes. Recently, the field of machine learning has been revolutionized by deep learning. In the context of image analysis, Convolutional Neural Networks are among the most successful deep learning algorithms as they can learn a hierarchy of spatial features at different layers of the network associated to increasing levels of abstraction, i.e., from raw pixel values to parts of objects (edges and corners), local shapes, up to complex structures in satellite and aerial images. In MASSIVE the most promising network architectures for pixel-wise mapping in remote sensing, Fully Convolutional Networks, will be adapted to map glacier extend and estimate glacier mass balance.

Introduction. Septic shock is a systemic inflammatory response syndrome associated with organ failures. Earlier clinical diagnosis would be of benefit to a decrease in the mortality rate. However, there is currently a lack of predictive biomarkers. The secretome is the set of proteins secreted by a cell, tissue, or organism at a given time and under certain conditions. The plasma secretome is easily accessible from biological fluids and represents a good opportunity to discover new biomarkers that can be studied with nontargeted “omic” strategies. Aims. To identify relevant deregulated proteins (DEP) in the secretome of a rat endotoxemic shock model. Methods. Endotoxemic shock was induced in rats by intravenous injection of lipopolysaccharides (LPS, S. enterica typhi, 0.5 mg/kg) and compared to controls (Ringer Lactate, iv). Under isoflurane anesthesia, carotid cannulation allowed mean arterial blood pressure (MAP) and heart rate (HR) monitoring and blood sampling at different time points (T0 and T50 or T0 and T90, with EDTA and protease inhibitor). Samples were prepared for large-scale tandem mass spectrometry (MS-MS) based on a label-free quantification to allow identification of the proteins deregulated upon endotoxemic conditions. A Gene Ontology (GO) analysis defined several clusters of biological processes (BP) in which the DEP are involved. Results. Ninety minutes after shock induction, the LPS group presents a reduction in MAP (-45%, p < 0.05 ) and increased lactate levels (+27.5%, p < 0.05 ) compared to the control group. Proteomic analyses revealed 10 and 33 DEP in the LPS group, respectively, at 50 and 90 minutes after LPS injection. At these time points, GO-BP showed alterations in pathways involved in oxidative stress response and coagulation. Conclusion. This study proposes an approach to identify relevant DEP in septic shock and brings new insights into the understanding of the secretome adaptations upon sepsis.

O-GlcNAcylation is a post-translational modification which affects approximately 5000 human proteins. Its involvement has been shown in many if not all biological processes. Variations in O-GlcNAcylation levels can be associated with the development of diseases. Deciphering the role of O-GlcNAcylation is an important issue to (i) understand its involvement in pathophysiological development and (ii) develop new therapeutic strategies to modulate O-GlcNAc levels. Over the past 30 years, despite the development of several approaches, knowledge of its role and regulation have remained limited. This review proposes an overview of the currently available tools to study O-GlcNAcylation and identify O-GlcNAcylated proteins. Briefly, we discuss pharmacological modulators, methods to study O-GlcNAcylation levels and approaches for O-GlcNAcylomic profiling. This review aims to contribute to a better understanding of the methods used to study O-GlcNAcylation and to promote efforts in the development of new strategies to explore this promising modification.

The synergistic combination of deep learning (DL) models and Earth observation (EO) promises significant advances to support the Sustainable Development Goals (SDGs). New developments and a plethora of applications are already changing the way humanity will face the challenges of our planet. This article reviews current DL approaches for EO data, along with their applications toward monitoring and achieving the SDGs most impacted by the rapid development of DL in EO. We systematically review case studies to achieve zero hunger, create sustainable cities, deliver tenure security, mitigate and adapt to climate change, and preserve biodiversity. Important societal, economic, and environmental implications are covered. Exciting times are coming when algorithms and Earth data can help in our endeavor to address the climate crisis and support more sustainable development.

Deep learning methods based on Fully convolution networks (FCNs) have shown an impressive progress in building outline delineation from very high resolution (VHR) remote sensing (RS) imagery. Common issues still exist in extracting precise building shapes and outlines, often resulting in irregular edges and over smoothed corners. In this paper, we use PolyMapper, a recently introduced deep-learning framework that is able to predict object outlines in a vector representation directly. We have introduced two main modifications to this baseline method. First, we introduce EffcientNet as backbone feature encoder to our network, which uses compound coefficient to scale up all dimensions of depth/width/resolution uniformly, to improve the processing speed with fewer parameters. Second, we integrate a boundary refinement block (BRB) to strengthen the boundary feature learning and to further improve the accuracy of corner prediction. The results demonstrate that the end-to-end learnable model is capable of delineating polygons of building outlines that closely approximate the structure of reference labels. Experiments on the crowdAI building instance segmentation datasets show that our model outperforms PolyMapper in all COCO metrics, for instance showing a 0.13 higher mean Average Precision (AP) value and a 0.60 higher mean Average Recall value. Also qualitative results show that our method segments building instances of various shapes more accurately.

One of the major challenges in precision viticulture in Europe is the detection and mapping of flavescence dorée (FD) grapevine disease to monitor and contain its spread. The lack of effective cures and the need for sustainable preventive measures are nowadays crucial issues. Insecticides and the plants uprooting are commonly employed to withhold disease infection, even if these solutions imply serious economic consequences and a strong environmental impact. The development of a rapid strategy to identify the disease is required to cover large portions of the crop and thus to limit damages in a time-effective way. This paper investigates the use of Unmanned Aerial Vehicles (UAVs), a cost-effective approach to early detection of diseased areas. We address this task with an object detection deep network, Faster R-CNN, instead of a traditional pixel-wise classifier. This work tests Faster R-CNN performance on this specific application through a comparative analysis with a pixel-wise classification algorithm (Random Forest). To take advantage of the full image resolution, the experimental analysis is performed using the original UAV imagery acquired in real conditions (instead of the derived orthomosaic). The first result of this paper is the definition of a new dataset for FD disease identification by UAV original imagery at the canopy scale. Moreover, we demonstrate the feasibility of applying Faster-R-CNN as a quasi-real-time alternative solution to semantic segmentation. The trained Faster-R-CNN achieved an average precision of 82% on the test set.

Fruit set, the transition from the flower to the growing fruit, is an essential step for fruit production. This process can be affected by environmental factors and by endogenous factors at the tree scale. The identification of these factors is important to improve cultivation practices and increase fruit set and fruit production. In this study, we focused on the identification of architectural factors affecting fruit set, with the mango tree, cv Cogshall, as a case study. The hypothesis is that topological, phenological and morphological characteristics of the flowering terminal growth unit affect its probability of fruit set. Two datasets describing exhaustively the vegetative and reproductive development of mango trees during two and three years, respectively, were analyzed. Generalized linear models were used to test the effects of the characteristics of the terminal growth units on their probability of fruit set. The effects of factors at a scale larger than the growth unit, such as the tree or the year were also tested. The date of burst of the terminal growth unit, i.e. its age at the time of flowering, its topological position, apical or lateral, and the number of inflorescences on the growth unit affected significantly its probability of fruit set during most of the years. The probability of fruit set was also affected by the year, by the tree and by the previous year tree yield, suggesting that factors at the tree scale and environmental factors were involved in fruit set. These results are useful to design cultivation practices aiming to improve mango tree fruit set. They will also be integrated in a model of mango fruit yield and quality at the tree scale to simulate more accurately fruit set and fruit production.