The past two decades have witnessed the great success of the algorithmic modeling framework advocated by Breiman et al. (2001). Nevertheless, the excellent prediction performance of these black-box models rely heavily on the availability of strong supervision, i.e. a large set of accurate and exact ground-truth labels. In practice, strong supervision can be unavailable or expensive, which calls for modeling techniques under weak supervision. In this comment, we summarize the key concepts in weakly supervised learning and discuss some recent developments in the field. Using algorithmic modeling alone under a weak supervision might lead to unstable and misleading results. A promising direction would be integrating the data modeling culture into such a framework.
Abstract Acute kidney injury (AKI) has considerably high morbidity and mortality but we do not have proper treatment for it. There is an urgent need to develop new prevention or treatment methods. Gut microbiota has a close connection with renal diseases and has become the new therapy target for AKI. In this study, we found the oral administration of the probiotic Limosilactobacillus reuteri had a prevention effect on the AKI induced by lipopolysaccharide (LPS). It reduced serum concentration of creatinine and urea nitrogen and protected the renal cells from necrosis and apoptosis. Meanwhile, L. reuteri improved the gut barrier function, which is destroyed in AKI, and modulated the gut microbiota and relevant metabolites. Compared with the LPS group, L. reuteri increased the proportion of Proteobacteria and reduced the proportion of Firmicutes, changing the overall structure of the gut microbiota. It also influenced the fecal metabolites and changed the metabolite pathways, such as tyrosine metabolism, pentose and glucuronate interconversions, galactose metabolism, purine metabolism, and insulin resistance. These results showed that L. reuteri is a potential therapy for AKI as it helps in sustaining the gut barrier integrity and modulating gut microbiota and related metabolites.
Abstract Background Heat dissipation relies on an intact cardiovascular system to dilate cutaneous blood vessels and increase cardiac output. However, the heart becomes a vulnerable organ and is susceptible to cardiac arrhythmias, functional failure, and focal myocardial necrosis in a hyperthermic heat-damaged state. In particular, people with cardiovascular dysfunction are at a much higher risk of exertional heat stroke (EHS). This study aimed to investigate and validate the cell signaling pathways and key genes associated with EHS by analyzing single-cell RNA sequencing (scRNA-seq) data from cardiac apical tissue of EHS rats. The findings are intended to elucidate the mechanisms underlying cardiac injury and to provide a theoretical basis for the early identification of biomarkers for cardiac injury in EHS. Results After exertional heat radiation, the heart's functionality was compromised. Annotation analysis revealed that the cell type and quantity did not differ between the EHS and control (CTL) groups. Cellchat analysis showed that the signal of EHS cardiac apex cells was enhanced in chemokine signaling pathway. The cardiac apical cells of the EHS group had the highest number of enriched genes in the oxidative stress pathway, according to GO/KEGG analysis of endothelial cells with the biggest proportion of cells. A total of 310 genes with changes in expression between the two groups were evaluated based on the Seurat-FindAllMarkers tools for all cell types. Of these, 18 genes with substantial variability were chosen for further verification. By using RT-qPCR verification, the expression differences of 12 genes were confirmed to be consistent with the above bioinformation analysis. Finally, Additional immunohistochemistry tests verified that Hspa8 and Hspe1 were up-regulated once more, while Id1, Ndufa4, and Cd36 were down-regulated. Conclusions The gene expression levels of Id1, Ndufa4, Cd36 were significantly reduced, and Hspa8, Hspe1 were significantly increased. These screened hypervariable genes play different roles in heat stress-induced mitochondrial and myocardial mechanical damage, protein misfolding, and they may become potential biomarkers in the mechanism of cardiac injury or keep an important link in the functional pathway of action described above.
Abstract Purpose This research was conducted to investigate the potential of short-chain fatty acids (SCFAs) in protecting organs from heat stress-induced injuries and gut microbiota modulation. Methods Sprague–Dawley rats were randomly assigned to various groups including a control group, a room temperature training group, a hyperthermia training group, SCFAs pretreatment group, and recipients received feces from the HT group. After strenuous training at high temperatures, the levels of plasma enzymes AST, ALT, BUN, and Cr were evaluated. The changes in gut microbiota and fecal metabolites were detected using 16S rRNA sequencing and GC–MS methods. Pathological examination of colon and liver tissues was conducted, and immunohistochemical techniques were employed to assess intestinal barrier function. Results The findings indicate that SCFAs hold the potential for mitigating liver and colon damage caused by heat stress. With the intervention of SCFAs, there were observable changes in the structure and metabolites of the intestinal microbiota, as well as improvements in intestinal barrier function. Further support for the benefits of SCFAs was found through fecal microbiota transplantation, which demonstrated that modified gut microbiota can effectively reduce organ damage. Conclusions This study provides evidence that SCFAs, as metabolites of the gut microbiota, have a valuable role to play in regulating gut health and mitigating the harmful effects of heat stress.
This paper applies the information fusion technology to tool monitoring. As one of the most important processing factor, the cutting tool and the tool wear directly influence size precision. Signals of cutting force and vibration are measured with multi-sensor. By using multi-sensor the drawbacks can be overcome, the multi-sensor information fusion mentioned in manufacture stands for extracting kinds of information from different sensors (especially for cutting force and vibration signal in this paper), making best use of all resources，according to certain criterion to judge the spatial and time redundancy , to make the system more comprehensive. Two data fusion methods, which are BP Neural Network and Wavelet Neural Network for predicting tool wear, and are debated. By the hybrid of BP and wavelet based neural network the cutting tool status inspection system is built so that the forecast of tool wear can be achieved. The results show experimentally two of these presented methods effectively implement tool wear monitoring and predicting.
Abstract Information about the spatial distribution of species lies at the heart of many important questions in ecology. Logistical limitations and collection biases, however, limit the availability of such data at ecologically relevant scales. Remotely sensed information can alleviate some of these concerns, but presents challenges associated with accurate species identification and limited availability of field data for validation, especially in high diversity ecosystems such as tropical forests. Recent advances in machine learning offer a promising and cost‐efficient approach for gathering a large amount of species distribution data from aerial photographs. Here, we propose a novel machine learning framework, artificial perceptual learning (APL), to tackle the problem of weakly supervised pixel‐level mapping of tree species in forests. Challenges arise from limited availability of ground labels for tree species, lack of precise segmentation of tree canopies and misalignment between visible canopies in the aerial images and stem locations associated with ground labels. The proposed APL framework addresses these challenges by constructing a workflow using state‐of‐the‐art machine learning algorithms. We develop and illustrate the proposed framework by implementing a fine‐grain mapping of three species, the palm Prestoea acuminata and the tree species Cecropia schreberiana and Manilkara bidentata , over a 5,000‐ha area of El Yunque National Forest in Puerto Rico. These large‐scale maps are based on unlabelled high‐resolution aerial images of unsegmented tree canopies. Misaligned ground‐based labels, available for <1% of these images, serve as the only weak supervision. APL performance is evaluated using ground‐based labels and high‐quality human segmentation using Amazon Mechanical Turk, and compared to a basic workflow that relies solely on labelled images. Receiver operating characteristic (ROC) curves and Intersection over Union (IoU) metrics demonstrate that APL substantially outperforms the basic workflow and attains human‐level cognitive economy, with 50‐fold time savings. For the palm and C. schreberiana , the APL framework has high pixelwise accuracy and IoU with reference to human segmentations. For M . bidentata , APL predictions are congruent with ground‐based labels. Our approach shows great potential for leveraging existing data from global forest plot networks coupled with aerial imagery to map tree species at ecologically meaningful spatial scales.
The global prevalence of allergic rhinitis (AR) remains high, posing challenges due to its chronic nature and propensity for recurrence. Gut microbiota dysbiosis contributes to immune dysregulation, impacting AR pathogenesis. Limosilactobacillus reuteri ( L. reuteri ) has great potential in regulating immune function to alleviate AR symptoms. However, the specific active components and mechanisms underlying its therapeutic effects in AR remain incompletely clarified. This study aimed to explore the potential mechanisms of L. reuteri and its metabolites in alleviating AR. The AR mouse model was constructed using ovalbumin (OVA). The analysis of hematoxylin–eosin staining (HE staining) and enzyme-linked immunosorbent assay (ELISA) suggested that L. reuteri alleviated nasal inflammation, suppressed aberrant Th2 immune responses, and modulated the balance of Treg and Th17 cytokines. The 16S rRNA sequencing and untargeted metabolic analysis revealed that L. reuteri restored gut microbiota composition and significantly increased the abundance of Ligilactobacillus and the metabolite luteolin (LO). Through ELISA and Western blotting analysis, LO treatment restored the Th1/Th2 and Treg/Th17 cytokine balance and suppressed the MAPK/STAT3 signaling pathway in AR mice. The study highlights LO as a key metabolite contributing to the anti-inflammatory effects of L. reuteri , suggesting potential avenues for future therapeutic strategies in AR management.
Learning conditional densities and identifying factors that influence the entire distribution are vital tasks in data-driven applications. Conventional approaches work mostly with summary statistics, and are hence inadequate for a comprehensive investigation. Recently, there have been developments on functional regression methods to model density curves as functional outcomes. A major challenge for developing such models lies in the inherent constraint of non-negativity and unit integral for the functional space of density outcomes. To overcome this fundamental issue, we propose Wasserstein Distributional Learning (WDL), a flexible density-on-scalar regression modeling framework that starts with the Wasserstein distance $W_2$ as a proper metric for the space of density outcomes. We then introduce a heterogeneous and flexible class of Semi-parametric Conditional Gaussian Mixture Models (SCGMM) as the model class $\mathfrak{F} \otimes \mathcal{T}$. The resulting metric space $(\mathfrak{F} \otimes \mathcal{T}, W_2)$ satisfies the required constraints and offers a dense and closed functional subspace. For fitting the proposed model, we further develop an efficient algorithm based on Majorization-Minimization optimization with boosted trees. Compared with methods in the previous literature, WDL better characterizes and uncovers the nonlinear dependence of the conditional densities, and their derived summary statistics. We demonstrate the effectiveness of the WDL framework through simulations and real-world applications.
There were defects like limited osteogenesis and fast drug release in traditional magnesium phosphate bone cement (MPC). In this study, we loaded icariin in a mesoporous nano silica container modified by polydopamine and then added it and citric acid into MPC (IHP-CA MPCs). The results indicate that IHP-CA MPCs have a long curing time, almost neutral pH value, excellent injectability, and compressive strength. In vitro experiments have shown that IHP-CA MPCs have good biocompatibility and bone promoting ability. These improvements provide feasible solutions and references for the clinical application of MPCs as implants.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
