6974 research outputs found
Sort by
Data for "Mean-field model for the bubble size distribution in coarsening wet foams"
No full textThis dataset contains the simulation data used in the linked output. The manuscript concerns predictions of the coarsening scaling state by a growth law we propose, which is validated against finite-element simulations of wet foams. Mean-field coarsening simulations are performed to verify that the derived scaling state is well behaved. Included in this dataset are python scripts to generate the manuscript's figures, in addition to simulation output, dump files, and code for the three-dimensional finite-element simulations and the mean-field simulations. For further details, please see the readme file
Modulated Convolutional Networks
Get PDFWhile the deep convolutional neural network (DCNN) has achieved overwhelming success in various vision tasks, its heavy computational and storage overhead hinders the practical use of resource-constrained devices. Recently, compressing DCNN models has attracted increasing attention, where binarization-based schemes have generated great research popularity due to their high compression rate. In this article, we propose modulated convolutional networks (MCNs) to obtain binarized DCNNs with high performance. We lead a new architecture in MCNs to efficiently fuse the multiple features and achieve a similar performance as the full-precision model. The calculation of MCNs is theoretically reformulated as a discrete optimization problem to build binarized DCNNs, for the first time, which jointly consider the filter loss, center loss, and softmax loss in a unified framework. Our MCNs are generic and can decompose full-precision filters in DCNNs, e.g., conventional DCNNs, VGG, AlexNet, ResNets, or Wide-ResNets, into a compact set of binarized filters which are optimized based on a projection function and a new updated rule during the backpropagation. Moreover, we propose modulation filters (M-Filters) to recover filters from binarized ones, which lead to a specific architecture to calculate the network model. Our proposed MCNs substantially reduce the storage cost of convolutional filters by a factor of 32 with a comparable performance to the full-precision counterparts, achieving much better performance than other state-of-the-art binarized models.</p
In-vitro investigation of mycoprotein:Opportunities for older adult nutrition
No full textDietary intake is a major determinant of later life course outcomes, with poor dietary intake linked with increased incidence of non-communicable diseases. Protein under-nutrition is a contributing factor to declining health in community dwelling older adults. Mycoprotein is a food ingredient produced from the fermentation of Fusarium venenatum, that is high in protein and fibre, has a complete amino acid profile as well as being low in fat(1). The aim of this study was to utilise in-vitro gastrointestinal digestion models to determine the extent of variability in the digestibility profile of mycoprotein when comparing a standard adult model, to the older adult model.Freezed dried Mycoprotein paste was subjected to in-vitro gastrointestinal digestion, using a computer controlled bioreactor system(2), utilising optimised conditions which simulated both adult and older adult gastrointestinal conditions(3). Following digestion, absorbable and non-absorbable protein fractions were separated and the protein content determined determined using total nitrogen analysis (Rapid Max N exceed, Elementar UK) and total amino acid analysis via ion exchange chromatography with post-column ninhydrin derivitisation (Hitachi LA8080, Avantor UK). All values were corrected using blank (protein free) digestions containing only digestive enzymes.After mass balance correction (for protein content of digestive enzymes and non-protein nitrogen), total in-vitro protein digestibility of mycoprotein was 65.0 % (SD 0.03 %) using standard digestion conditions in triplicate. For conditions that simulated older adult digestion, total protein digestibility was reduced by 3.4 %. Digestibility of indispensable amino acids between the two models, were non-significant (p > 0.05). Previously published trials have shown that for plant based feedstocks, protein digestibility reduced by up to 20 % using older adult conditions.As the global food system evolves to meet the nutritional needs of a growing population and challenges of Net Zero, there is potential for high protein plant based alternatives to meet the nutritional needs of an ageing population. This work has demonstrated that the digestion of mycoprotein is largely unaffected by the changing gastrointestinal conditions of older adults and therefore could play an important role in future nutritional support strategies for older adults
Exploring the Capability of Earth Observation Data and a New ‘4S’ Framework for Disaster Risk Reduction:An Experience from July 2022 Flash Floods in Amarnath Valley, India
No full textThe holy cave shrine of Amarnath is thronged annually by over three lakh devotees. However, the cryogenic-sensitive region is increasingly becoming vulnerable to anomalies in precipitation events and rising anthropogenic footprints. On the evening of 8 July 2022, a highly localized extreme rainfall event took place leading to a short-lived flash flood along with unsorted debris flow in the Amarnath valley, surmounting heavy loss of lives and local livelihoods. This article uses Earth Observation (EO) datasets to capture and understand the implications of using such data applications in mitigating disasters for remote and inaccessible areas. Despite the valuable insights provided by EO data, their applicability is often restricted by the temporal limitations, particularly those derived from open-source radar and optical satellite imagery, which are frequently incapable of capturing ephemeral or rapidly evolving phenomena. Some meaningful information about the present study was captured with the use of GPM (IMERG) satellite-based rainfall data, while others failed to gauge the situation. Eight topographical parameters have been examined to understand the local factors contributing to flash flood conditions in the Amarnath watershed. An AHP-based flash flood susceptibility zonation (FFSZ) was derived using Google Earth Engine (GEE) along with an interactive user interface was developed for visualization of the computed parameters. The FFSZ contains five classes with their areal percentages: Very Low (20.03%), Low (19.69%), Moderate (20.43%), High (20.14%) and Very High (19.71%) respectively. Our findings suggest the need for more ground-based automated weather stations (AWS) complementing satellite-based EO systems' limitations for providing high-precision regular interval observation. Finally, we propose a new ‘4S’ framework, namely, ‘source’, ‘setting’, ‘susceptibility’ and ‘solution’ for flash flood risk assessment. This framework has also been discussed in complementary to a cross-sectoral interface containing ‘science-governance-disaster risk reduction (DRR)-society’ aspects alongside major targets based on Global Goals (UN SDGs) and India’s national DRR agenda points.</p
Body condition score and weight are effective targeted selective treatment indicators for gastrointestinal nematodes in premating ewes
No full textBackground: Ewe reproductive performance is key for flock profitability but can be impacted upon by gastrointestinal nematode (GIN) infections. Farmers commonly deworm ewes pre‐mating, yet concerns regarding anthelmintic resistance increase pressure to reduce wormer use. Methods: This study retrospectively assessed pre‐mating targeted selective treatment (TST) indicators in a flock of 354 ewes split into anthelmintic treatment and control groups. Conway–Maxwell–Poisson and binomial regression analyses were used to identify TST indicators associated with reproductive performance. Results: There was no significant difference in overall scanned litter size between treated and untreated ewes. However, ewes with a body condition score (BCS) less than 3 or a weight less than 60 kg that were not dewormed had significantly lower litter sizes (mean 1.2 and 1.41, respectively) than those treated (1.94 and 1.8, respectively) or with higher condition/weight regardless of worming status (p < 0.05). Limitations: Parasite diagnostics were not undertaken on all individual ewes, and the study assumes that differences identified in the reproductive performance of treated and untreated groups were due to GIN infection. Conclusion: These findings suggest that BCS and weight are effective indicators for implementing TST pre‐mating, enabling reduced anthelmintic use without compromising reproductive outcomes
The Urinary Proteome Differs with the Presence and Type of Breast Cancer
Get PDFDespite advancements in screening and treatment, the incidence of breast cancer (BC) and associated mortality are projected to increase. Therefore, developing a companion diagnostic for BC remains important. Herein, we explore the urinary proteome for biomarkers of BC: 130 urine samples from (1) newly diagnosed breast cancer (BC), n = 46, (2) benign breast disease (BBD), n = 36, (3) symptom control (SC), n = 30, and (4) healthy control (HC), n = 18. The BC class included preinvasive: ductal carcinoma in situ (DCIS) (n = 3), invasive ductal carcinoma (IDC) (n = 23), and IDC accompanied by DCIS (n = 8) classes. Protein profiling was performed using ThermoScientific ProteomeDiscoverer and analyzed using MetaboAnalyst v6.0, DAVID, and STRING v12.0. Analyses identified 346 significantly (p < 0.05) differentially expressed proteins (DEP) across BC, BBD, SC, and HC. Multivariate Receiver Operating Characteristic curves (five proteins) suggested Area Under the Curve values of 0.985, 0.989, and 0.999 distinguishing BC from BBD, SC, and HC, respectively. DEP elevated in BC included beta-glucuronidase isoform 1, fibrinogen gamma chain, alpha-actinin-1, peptidase inhibitor 16, cysteine-rich C-terminal protein 1 isoform X1, guanine nucleotide-binding protein G(I)/G(S)/G(T) subunit beta-1, vascular cell adhesion protein 1, ATP-dependent translocase ABCB1, and tumor protein p63-regulated gene 1 isoform X1. BC types were differentiated based on calpain-2 and cystatin-C expression (p < 0.05). Thus, BC has distinct urinary–protein profiles based on clinical diagnosis, which could be used in real-time noninvasive BC monitoring.</p