Search

The discovery of new methods to control porosity and microarchitecture may lead to the refinement of carbon materials from lignocellulose as advanced functional materials. However, the high ash content on the surface of lignocellulosic biomass reduces the surface area and adsorption properties of the activated carbon. This study presents a novel approach, using a deashing post-treatment as the pore generator, to increase the quality of the activated carbon. The micropore capacity was improved by deashing post-treatment with distilled water, where 80% of the total pore ratio of the activated carbon was occupied with micropores. Ultrasonic treatment was able to penetrate deeper into the structure of coconut shell activated carbon, creating cavities and pores, thus increasing the surface area. Understanding the effects of these new controlling methods on pore refinement can elucidate the microporous fabrication of other activated carbons from high ash-content lignocellulosic biomass.

The invention of spindleless lathe technology has enabled veneers to be produced from small logs, such as logs from short rotation Hevea plantations, with low recovery loss. However, for structural laminated products, such as laminated veneer lumber (LVL), manufacturers are highly selective regarding the veneers for their conventional production. During the peeling process of small logs (< 18 cm), deeper and higher frequency of lathe checks were induced on veneer surface compared to the common log size used (> 30 cm). In this study, spindleless rotary-peeled veneers made from small rubber logs were processed into LVL using different lamination pressures: 7, 8, 9, and 10 kgf/cm2. The effects of lamination pressures on the physical and mechanical properties of the produced LVL were evaluated. Based on the findings, the specific gravity increased from 0.73 to 0.83 with increased lamination pressure. In terms of mechanical properties, all the values increased with lamination pressure, but with a sudden drop with 10 kgf/cm2. Understanding the effect of lamination pressure on the physical and mechanical properties can shed light on optimizing the usage of spindleless rotary-peeled veneers from small logs for the production LVL and other lamination products.

In this study, the degradation efficiency and electricity generation of the azo dyes affected by the functional groups and molecular structure in a solar photocatalytic fuel cell (PFC) system were investigated and discussed in detail. Four different azo dyes such as, Acid Orange 7 (AO7), Acid Red 18 (AR18), Reactive Black 5 (RB5), Reactive Red 120 (RR120) with different molecular structure were evaluated. The degradation efficiency of AO7, AR18, RB5 and RR120 achieved 5.6 ± 0.3%, 11.1 ± 0.6%, 41.9 ± 0.9% and 52.1 ± 1.3%, respectively, after 6 h irradiated under solar light. In addition, the maximum power density, P(max) for AO7, AR18, RB5 and RR120 was 0.0269 ± 0.01, 0.111 ± 0.03, 1.665 ± 0.67 and 4.806 ± 1.79 mW cm(−2), respectively. Meanwhile, the concentration of COD for AO7, AR18, RB5 and RR120 reduced to 16 ± 0.1, 10 ± 0.3, 7 ± 0.6 and 3 ± 0.9 mg L(−1), respectively. The concentration ratio of benzene / naphthalene, benzene / azo bond and naphthalene / azo bond, respectively, was analyzed to investigate the impact of the functional groups over photodegradation of the azo dyes in PFC. Electron releasing groups (-OH and –NH(2)) and electron withdrawing groups (-SO(3)Na) which attached to the naphthalene or benzene ring also played a pivotal role in the degradation mechanism.

Summary Background Deep learning is a novel machine learning technique that has been shown to be as effective as human graders in detecting diabetic retinopathy from fundus photographs. We used a cost-minimisation analysis to evaluate the potential savings of two deep learning approaches as compared with the current human assessment: a semi-automated deep learning model as a triage filter before secondary human assessment; and a fully automated deep learning model without human assessment. Methods In this economic analysis modelling study, using 39 006 consecutive patients with diabetes in a national diabetic retinopathy screening programme in Singapore in 2015, we used a decision tree model and TreeAge Pro to compare the actual cost of screening this cohort with human graders against the simulated cost for semi-automated and fully automated screening models. Model parameters included diabetic retinopathy prevalence rates, diabetic retinopathy screening costs under each screening model, cost of medical consultation, and diagnostic performance (ie, sensitivity and specificity). The primary outcome was total cost for each screening model. Deterministic sensitivity analyses were done to gauge the sensitivity of the results to key model assumptions. Findings From the health system perspective, the semi-automated screening model was the least expensive of the three models, at US$62 per patient per year. The fully automated model was $66 per patient per year, and the human assessment model was $77 per patient per year. The savings to the Singapore health system associated with switching to the semi-automated model are estimated to be $489 000, which is roughly 20% of the current annual screening cost. By 2050, Singapore is projected to have 1 million people with diabetes; at this time, the estimated annual savings would be $15 million. Interpretation This study provides a strong economic rationale for using deep learning systems as an assistive tool to screen for diabetic retinopathy. Funding Ministry of Health, Singapore.

Although deep learning for Diabetic Retinopathy (DR) screening has shown great success in achieving clinically acceptable accuracy for referable versus non-referable DR, there remains a need to provide more fine-grained grading of the DR severity level as well as automated segmentation of lesions (if any) in the retina images. We observe that the DR severity level of an image is dependent on the presence of different types of lesions and their prevalence. In this work, we adopt a multi-task learning approach to perform the DR grading and lesion segmentation tasks. In light of the lack of lesion segmentation mask ground-truths, we further propose a semi-supervised learning process to obtain the segmentation masks for the various datasets. Experiments results on publicly available datasets and a real world dataset obtained from population screening demonstrate the effectiveness of the multi-task solution over state-of-the-art networks.

The present research was conducted to systematically review existing research that examined the relationships of the aspects of self-construal and social network sites (SNS) usages. A total of 12 research articles met the inclusion criteria for the present review. The reviewed research articles mainly supported the significant relationship of the interdependent self-construal and SNS-related outcomes. The present review highlighted that the reviewed relationships differed. Some findings supported the direct effect of self-construal on SNS-related outcomes, whereas some supported the indirect effects of intervening variables on these relationships. The reviewed findings supported the influence of self-construal on cognition, emotion, and motivation. Implications of the present systematic review were discussed in the manuscript.

Background Aortic dissection is a life-threatening syndrome that sometimes requires emergency intervention, and endovascular aortic aneurysm repair (EVAR) is a treatment option. Long-term image follow-up is also required for patients after EVAR due to possible complications. Case presentation We present the case of a 73-year-old male with underlying chronic renal disease diagnosed with a type A aortic dissection who underwent EVAR. Four-dimensional (three spatial dimensions combined with time) phase-contrast magnetic resonance imaging (4D PC-MRI) was performed during regular follow-up in preference to contrast-enhanced computed tomography or simple MRI while taking his poor renal function into consideration. Conclusions We considered this preferable given his issues with renal function.

Importance: Due to the evolving variants of coronavirus disease 2019 (COVID-19), it is important to understand the relationship between the disease condition and socioeconomic, demographic, and health indicators across regions. Background: Studies examining the relationships between infectious disease and socioeconomic variables are not yet well established. Design: A total of 3042 counties in the United States are included as the observation unit in the study. Two outcome variables employed in the study are the control of disease spread and infection prevalence rates in each county. Method: Data are submitted to quantile regression, hierarchical regression, and random forest analyses to understand the extent to which health outcomes are affected by demographics, socioeconomics, and health indicators. Results: Counties with better control of the disease spread tend to have lower infection rates, and vice versa. When measuring different outcome variables, the common risk factors for COVID-19 with a 5% level of statistical significance include employment ratio, female labor ratio, young population ratio, and residents’ average health risk factors, while protective factors include land size, housing value, travel time to work, female population ratio, and ratio of residents who identify themselves as mixed race. Conclusions: The implications of the findings are that the ability to maintain social distancing and personal hygiene habits are crucial in deterring disease transmission and lowering incidence rates, especially in the early stage of disease formation. Relevant authorities should identify preventive factors and take early actions to fight infectious diseases in the future.

Myriad neuropsychiatric disorders are due to dopamine dysfunction. However, understanding these disorders is limited by our ability to measure dopamine storage and release. Fluorescent false neurotransmitters (FFNs), small-molecule dyes that co-transit through the synaptic vesicle cycle, have allowed us to image dopamine in cell culture and acute brain slice, but in vivo microscopy is constrained by the biopenetrance of light. Here, we adapt FFNs into magnetic resonance false neurotransmitters (MFNs). The design principles guiding MFNs are (1) the molecule is a valid false neurotransmitter and (2) it has a 19F-substituent near a pH-sensing functional group, which (3) has pKa close to 6 so that the probe within vesicles is protonated. We demonstrate that MFN103 meets these criteria. While a magnetic resonance spectroscopy (MRS) signal was too low for measurement in vivo with the current technology, in principle, MFNs can quantify neurotransmitters within and without synaptic vesicles, which may underlie noninvasive in vivo analysis of dopamine neurotransmission.

Latex production from Hevea brasiliensis rubber tree is the second most important commodity in Malaysia, but this industry is threatened by the white root rot disease (WRD) caused by Rigidoporus microporus that leads to considerable latex yield loss and tree death. This study aimed to characterize and compare the virulence of five R. microporus isolates obtained from infected rubber trees located at different states in Malaysia. These isolates were subjected to morphological and molecular characterization for species confirmation and pathogenicity test for the determination of virulence level. BLAST search showed that the ITS sequences of all the pathogen isolates were 99% identical to R. microporus isolate SEG (accession number: MG199553) from Malaysia. The pathogenicity test of R. microporus isolates conducted in a nursery with 24 seedlings per isolate showed that isolate RL21 from Sarawak has developed the most severe above- and below-ground symptoms of WRD on the rubber clone RRIM600 as host. Six months after being infected with R. microporus, RL21 was evaluated with the highest average of disease severity index of 80.52% for above- and below-ground symptoms, followed by RL22 (68.65%), RL20 (66.04%), RL26 (54.38%), and RL25 (43.13%). The in vitro growth condition tests showed that isolate RL21 of R. microporus has optimum growth at 25–30 °C, with the preference of weakly acidic to neutral environments (pH 6–7). This study revealed that different virulence levels are possessed among different R. microporus isolates even though they were isolated from the same host species under the same climate region. Taken together, field evaluation through visual observation and laboratory assays have led to screening of the most virulent isolate. Determination of the most virulent isolate in the present study is vital and shall be taken into consideration for the selection of suitable pathogen isolate in the development of more effective control measures in combating tenacious R. microporus.

Deep learning-based models are developed to automatically detect if a retina image is `referable' in diabetic retinopathy (DR) screening. However, their classification accuracy degrades as the input images distributionally shift from their training distribution. Further, even if the input is not a retina image, a standard DR classifier produces a high confident prediction that the image is `referable'. Our paper presents a Dirichlet Prior Network-based framework to address this issue. It utilizes an out-of-distribution (OOD) detector model and a DR classification model to improve generalizability by identifying OOD images. Experiments on real-world datasets indicate that the proposed framework can eliminate the unknown non-retina images and identify the distributionally shifted retina images for human intervention., Accepted at IEEE ICIP 2021

The aim of this study was to select the optimal conditions for the carbonization process followed by surface modification treatment with sodium hydroxide (NaOH) to obtain a highly microporous activated carbon structure derived from palm kernel shells (PKS) and coconut shells (CS). The effects of the carbonization temperature and NaOH concentration on the physiochemical properties, adsorption capability, specific surface area, surface morphology, and surface chemistry of PKS and CS were evaluated in this study. The results show that surface-modified activated carbons presented higher surface area values (CS: 356.87 m2 g−1, PKS: 427.64 m2 g−1), smaller pore size (CS: 2.24 nm, PKS: 1.99 nm), and larger pore volume (CS: 0.34 cm3 g−1, PKS: 0.30 cm3 g−1) than the untreated activated carbon, demonstrating that the NaOH surface modification was efficient enough to improve the surface characteristics of the activated carbon. Moreover, surface modification via 25% NaOH greatly increases the active functional group of activated carbon, thereby directly increasing the adsorption capability of activated carbon (CS: 527.44 mg g−1, PKS: 627.03 mg g−1). By applying the NaOH post-treatment as the ultimate surface modification technique to the activated carbon derived from PKS and CS, a highly microporous structure was produced.

Simple Summary Blastocystis spp. are unicellular parasites that infect the gastrointestinal tract of humans and animals. Their occurrence in the environment had been detected in water sources, thus causing contamination. The presence of the parasites in humans, animals and environmental sources in Asia were reviewed according to countries in Asia, different categories of human and animal populations, and environmental sources including water samples, food and ambient air. The coexistence of the parasites poses a public health concern as the parasites are commonly found in most studies. Hence, there is a growing interest in the study of Blastocystis spp. Due to the isolation of Blastocystis spp. from living and non-living sources, a collaborative, multisectoral and transdisciplinary approach known as One Health is proposed for future study of Blastocystis spp. in order to achieve optimal health outcomes through the recognition of interconnection between people, animals and their shared environment. Abstract Blastocystis spp. are controversial unicellular protists that inhabit the gastrointestinal tract of humans and a wide range of animals worldwide. This review provides an overview of the prevalence and distribution of Blastocystis spp. and their subtypes throughout Asia. Research articles reporting on the presence of Blastocystis spp. in locations within Asia, between 1 January 2010, and 10 May 2021, were obtained from Scopus, PubMed, and Google Scholar. In 427 articles, the prevalence of Blastocystis spp. in 31 countries within the last decade was revealed. Isolates were found in humans, various mammals, birds, reptiles, insects, water sources, vegetables, and ambient air. Prevalence of Blastocystis spp. varied widely across host categories. Subtypes identified throughout Asia were STs 1–14, and ST18–22 (novel subtypes). ST1, ST2, ST3, ST4 were the most frequently isolated in humans; ST5 in pigs; ST10 and ST14 in goats, sheep, and cattle; and ST6 and ST7 in chickens. ST1 and ST3 were most common in water samples. ST1, ST2, ST3, ST4, ST5 and ST6 were shared by humans, animals, and water sources. There is a growing interest in the study of Blastocystis spp. and their subtypes in Asia. Due to the isolation of Blastocystis spp. from biotic and abiotic sources in Asia, the application of the One Health (OH) approach to the study of Blastocystis spp. is proposed for improved perception of this organism.

Unproductive young rubber trees (15 years old) with smaller diameters (15 to 18 cm) compared to conventional rubber logs, harvested at the age of 25 years old, were selected for the production of laminated panels. Spindleless rotary veneer peeling was applied to logs from short-rotation rubber forest plantations to produce veneers for structural purposes. This raises questions about the utilization of these small-diameter logs with respect to its effect on the quality of veneer and laminated panels produced. This study examines the effect of the glue spread rates on the physical and mechanical properties of rubberwood laminated veneer lumber (LVL). Analysis of variance shows that the application of a 280 g/m2 glue spread rate significantly improved the density, water absorption and dimensional stability of rubberwood LVL. The mechanical properties of rubberwood LVL produced with a 200 g/m2 glue spread rate met the minimum requirement for the 2.1E-3100F stress class, 91.05 MPa for the modulus of rupture in the flatwise direction and 50.23 MPa for compressive strength parallel to the longitudinal axis. The modulus of elasticity in the flatwise direction of 11,189.55 MPa reached the minimum requirement for the 1.5E-2250F stress class.

Interventions engaging community health workers (CHW) for diabetes management aim to improve diabetes care and self-management behaviors among patients. We evaluated the effects of nonsurgical periodontal treatment (NSPT) with the CHW strategy on oral self-care behaviors, periodontal status and oral health-related quality of life (OHQoL) in patients with type 2 diabetes mellitus (T2DM). The participants were randomly assigned to experimental (EG, n = 35) and control (CG, n = 33) groups. All participants received NSPT, whereas the patients in the EG also received one-on-one 30 min lessons from a CHW over 4 weeks. The EG exhibited greater improvement in the probing pocket depth (β = −0.2, effect size [ES] = 0.61) and clinical attachment level (β = −0.2, ES = 0.59) at 1-month follow-up than the CG did. The ES increased over the 1-, 3- and 6-month follow-ups, indicating an increase in OHQoL (ES = 0.19, 0.60, and 0.62, respectively) in the EG. The patients in the EG were more likely to change their oral self-care behaviors than patients in the CG were. The NSPT with CHW strategy had a positive effect on 1-month periodontal treatment outcomes, long-term OHQoL and oral self-care behaviors in patients with T2DM.

Background Clinical trials have demonstrated that initiating oral anti-diabetic drugs (OADs) significantly reduce glycated hemoglobin (HbA1c) levels. However, variability in lifestyle modifications and OAD adherence impact on their actual effect on glycemic control. Furthermore, evidence on dose adjustments and discontinuation of OAD on HbA1c is lacking. This study aims to use real-world data to determine the effect of OAD initiation, up-titration, down-titration, and discontinuation on HbA1c levels, among Asian patients managed in primary care. Methods A retrospective cohort study over a 5-year period, from Jan 2015 to Dec 2019 was conducted on a cohort of multi-ethnic adult Asian patients with clinical diagnosis of type 2 diabetes mellitus (T2DM) managed by a network of primary care clinics in Singapore. Nine OADs from five different classes (biguanides, sulphonyurea, dipeptidyl peptidase-4 [DPP-4] inhibitors, sodium-glucose cotransporter-2 [SGLT-2] inhibitors, and alpha-glucosidase inhibitors) were evaluated. Patients were grouped into “No OAD”, “Non-titrators,” and “Titrators” cohorts based on prescribing patterns. For the “Titrators” cohort, the various OAD titrations were identified. Subsequently, a descriptive analysis of HbA1c values before and after each titration was performed to compute a mean difference for each unique titration identified. Results Among the cohort of 57,910 patients, 43,338 of them had at least one OAD titration, with a total of 76,990 pairs of HbA1c values associated with an OAD titration. There were a total of 206 unique OAD titrations. Overall, initiation of OADs resulted in a reduction of HbA1c by 3 to 12 mmol/mol (0.3 to 1.1%), respectively. These results were slightly lower than those reported in clinical trials of 6 to 14 mmol/mol (0.5 to 1.25%). The change of HbA1c levels due to up-titration, down-titration, and discontinuation were −1 to −8 mmol/mol (−0.1 to −0.7%), +1 to 7 mmol/mol (+0.1 to +0.6%), and +2 to 11 mmol/mol (+0.2 to +1.0%), respectively. The HbA1c lowering effect of initiating newer OADs, namely DPP-4 inhibitors and SGLT-2 inhibitors was 8 to 11 mmol/mol (0.7 to 0.9%) and 7 to 11 mmol/mol (0.6 to 1.0%), respectively. Conclusion The real-world data on Asians with T2DM in this study show that the magnitudes of OAD initiation and dose titration are marginally lower than the results from clinical trials. During shared decision-making in selecting treatment options, the results enable physicians to communicate realistic expectation of the effect of oral medications on the glycemic control of their patients in primary care.

Patient similarity analytics has emerged as an essential tool to identify cohorts of patients who have similar clinical characteristics to some specific patient of interest. In this study, we propose a patient similarity measure called D3K that incorporates domain knowledge and data-driven insights. Using the electronic health records (EHRs) of 169,434 patients with either diabetes, hypertension or dyslipidaemia (DHL), we construct patient feature vectors containing demographics, vital signs, laboratory test results, and prescribed medications. We discretize the variables of interest into various bins based on domain knowledge and make the patient similarity computation to be aligned with clinical guidelines. Key findings from this study are: (1) D3K outperforms baseline approaches in all seven sub-cohorts, (2) our domain knowledge-based binning strategy outperformed the traditional percentile-based binning in all seven sub-cohorts, (3) there is substantial agreement between D3K and physicians (κ = 0.746), indicating that D3K can be applied to facilitate shared decision making. This is the first study to use patient similarity analytics on a cardiometabolic syndrome-related dataset sourced from medical institutions in Singapore. We consider patient similarity among patient cohorts with the same medical conditions to develop localized models for personalized decision support to improve the outcomes of a target patient.

Background This research aimed to demonstrate a significant suppressor effect between Facebook usages (Facebook intensity and Facebook addiction) on obsessive beliefs (threat, perfectionism, inflated responsibility, control of thoughts). Subjects and Methods Participants were 204 university students who completed self-reported measures of Facebook intensity, Facebook addiction, and obsessive beliefs. To examine the significance of the suppressor effect, the conceptual models were analyzed using GSCA Pro. Results Results indicated that the suppressor effect was not significant. Facebook intensity, which refers to regular Facebook use, did not significantly predict the facets of obsessive beliefs. Facebook addiction, which represents the maladaptive use of Facebook, significantly predicted these facets. Conclusions Results indicated that Facebook intensity did not increase the relationship between Facebook addiction and obsessive beliefs. Facebook addiction increases the saliency of obsessive beliefs, which possibly increases the risk of OCD. Acknowledgements This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Due to the corona (COVID-19) pandemic, several countries are currently conducting non-face-to-face education. Therefore, teachers of nursing colleges have been carrying out emergency remote education. This study developed a questionnaire to understand the status of Emergency Remote Learning (ERL) in nursing education internationally, translated it into 7 languages, and distributed it to 18 countries. A total of 328 nursing educators responded, and the most often used online methods were Social networking technology such as Facebook, Google+ and Video sharing platform such as YouTube. The ERL applied to nursing education was positively evaluated as 3.59 out of 5. The results of the study show that during the two semesters nursing college professors have well adapted to this unprecedent crisis of teaching. The world after COVID-19 has become a completely different place, and nursing education should be prepared for ‘untact’ education.

The growth of SrRuO$_3$ (SRO) thin film with high-crystallinity and low residual resistivity (RR) is essential to explore its intrinsic properties. Here, utilizing the adsorption-controlled growth technique, the growth condition of initial SrO layer on TiO$_2$-terminated SrTiO$_3$ (STO) (001) substrate was found to be crucial for achieving a low RR in the resulting SRO film grown afterward. The optimized initial SrO layer shows a $c$(2 x 2) superstructure that was characterized by electron diffraction, and a series of SRO films with different thicknesses ($t$s) were then grown. The resulting SRO films exhibit excellent crystallinity with orthorhombic-phase down to $t \approx$ 4.3 nm, which was confirmed by high resolution X-ray measurements. From azimuthal X-ray scan for SRO orthorhombic (021) reflection, we uncover four structural domains with a dominant domain of orthorhombic SRO [001] along cubic STO [010] direction. The dominant domain population depends on $t$, STO miscut angle (${\alpha}$), and miscut direction (${\beta}$), giving a volume fraction of about 92 $\%$ for $t \approx$ 26.6 nm and (${\alpha}$, ${\beta}$) ~ (0.14$^{\rm o}$, 5$^{\rm o}$). On the other hand, metallic and ferromagnetic properties were well preserved down to $t \approx$ 1.2 nm. Residual resistivity ratio (RRR = ${\rho}$(300 K)/${\rho}$(5 K)) reduces from 77.1 for $t \approx$ 28.5 nm to 2.5 for $t \approx$ 1.2 nm, while ${\rho}$(5 K) increases from 2.5 $\mu\Omega$cm for $t \approx$ 28.5 nm to 131.0 $\mu\Omega$cm for $t \approx$ 1.2 nm. The ferromagnetic onset temperature ($T_c\prime$) of around 151 K remains nearly unchanged down to $t \approx$ 9.0 nm and decreases to 90 K for $t \approx$ 1.2 nm. Our finding thus provides a practical guideline to achieve high crystallinity and low RR in ultra-thin SRO films by simply adjusting the growth of initial SrO layer., Comment: 22 pages, 7 figures

This research examined the relationship between social network site (SNS) intensity, SNS addiction, and the severity of obsessive-compulsive disorder (OCD), alongside its factors of obsession and compulsion. The overlap of SNS intensity and SNS addiction was controlled in the study to predict the measured severity of OCD. In this study, 204 Malaysian undergraduate students were recruited to complete the revised Facebook Intensity Scale, the revised Bergen Facebook Addiction Scale, and the self-report version of the Yale-Brown Obsessive-Compulsive Scale. The overlap of SNS intensity and SNS addiction was supported by their significant positive correlation. Furthermore, SNS addiction significantly correlated with the measured OCD and its corresponding factors. The hierarchical regression analysis revealed that the entry of SNS intensity enhanced the facilitative effect of SNS addiction on OCD and its factors. Therefore, the role of SNS intensity as a suppressor was supported. In the same regression model, SNS intensity predicted the severity of OCD and its compulsion subscale negatively. Implications and directions for future research were also discussed in this manuscript.

Microblogs have become the preferred means of communication for people to share information and feelings, especially for fast evolving events. Understanding the emotional reactions of people allows decision makers to formulate policies that are likely to be more well-received by the public and hence better accepted especially during policy implementation. However, uncovering the topics and emotions related to an event over time is a challenge due to the short and noisy nature of microblogs. This work proposes a weakly supervised learning approach to learn coherent topics and the corresponding emotional reactions as an event unfolds. We summarize the event by giving the representative microblogs and the emotion distributions associated with the topics over time. Experiments on multiple real-world event datasets demonstrate the effectiveness of the proposed approach over existing solutions.

The success of microbial termiticides in controlling termites depends on the ability of microbes to grow in different media and the functionality of the microbes as a resistant barrier or toxic bait. This study was conducted to understand the mortality rate and behaviour changes of the subterranean termite Coptotermes curvignathus Holmgren introduced with different concentrations of Serratia marcescens strain LGMS 1 and Pseudomonas aeruginosa strain LGMS 3 using wood and soil as bacterial transfer medium. In general, higher concentration of bacteria in soil caused a reduction in tunnelling activity and wood consumption and an increase in mortality. However, application on wood revealed a different outcome. Wood treated with S. marcescens of 10 6 CFU ml −1 concentration proved to be more efficient as bait than higher concentration applications as it caused a high mortality rate while still highly palatable for termites. Wood or soil treated with S. marcescens concentration higher than 10 9 CFU ml −1 creates a high toxicity and repellent barrier for termites. Pseudomonas aeruginosa of 10 9 CFU ml −1 concentrations applied on wood served as a slow-acting toxic bait. However, the ability for S. marcescens and P. aeruginosa to survive on wood is low, which made the bait unable to retain a useful level of toxicity for a long period of time and frequent reapplication is needed.

Background Sepsis causes death and morbidity in young infants. Globally, an estimated 1.3 – 3.9 million young infants experience sepsis and 400,000 – 700,000 die from sepsis-related conditions annually. Even though there have been significant progress over the past twenty years in reducing young infant mortality, sepsis currently accounts for up to 15% of all young infant deaths. A thorough understanding of young infant sepsis can inform strategies that span prevention, diagnosis and intervention for young infant sepsis. Objectives We aimed to perform a systematic review and meta-analysis to investigate the case fatality rates (CFRs) among young infants less than 90 days with sepsis globally. Methods We used the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) 2009 guidelines. We searched PubMed, Cochrane Central, Embase and Web of Science for randomized clinical trials and observational studies in English language, published between 2010 to 2019. Studies involving young infants less than 90 days old with sepsis and reported CFRs were included. We obtained pooled CFRs estimates using the random effects model. Additional stratifications by gestation, birth weight, onset of sepsis (early onset was defined as Results Among 6314 articles screened, 240 studies with a total of 437,796 patients met the inclusion criteria and were included in our analysis. 99 came from high income countries, 44 from upper middle income countries, 82 from lower middle income countries, 6 from low income countries and 9 were conducted in multiple countries. Overall, the pooled CFR was 0.18 (95% CI, 0.17–0.19). The CFR was the highest in low income countries (0.25 [95% CI, 0.07–0.43]), followed by lower middle (0.24 [95% CI, 0.21–0.26]), upper middle (0.21 [95% CI, 0.18–0.24]) and lastly high income countries (0.12 [95% CI, 0.11–0.13]). Other factors associated with higher CFRs included prematurity (0.23 [95% CI, 0.19–0.26] vs term CFR 0.10 [95% CI, 0.08–0.13]), low birth weight (0.21 [95% CI, 0.19–0.24] vs normal birth weight 0.19 [95% CI, 0.18–0.20]), early onset sepsis (0.20 [95% CI, 0.17–0.24] vs combined (0.16 [95% CI, 0.14–0.18]) and hospital acquired infection (0.23 [95% CI, 0.17–0.30] vs community acquired infections 0.21 [95% CI, 0.10–0.33]). Time trend analysis showed higher CFRs in the low income countries than the middle and high income countries. A decreasing trend in CFRs over time was observed in high and upper middle income countries, as compared to an increasing trend in lower middle and low income countries. Conclusions While we saw a declining trend of young infant sepsis CFRs among high and upper middle income countries across the years, the increasing trend amongst lower middle and low income countries highlights a disparity in infant sepsis outcomes based on resource availability. We highlight specific vulnerable patient populations that should be further studied in order to reduce the global burden of young infant sepsis.

Link prediction is a fundamental challenge in network science. Among various methods, local similarity indices are widely used for their high cost-performance. However, the performance is less robust: for some networks local indices are highly competitive to state-of-the-art algorithms while for some other networks they are very poor. Inspired by techniques developed for recommender systems, we propose an enhancement framework for local indices based on collaborative filtering (CF). Considering the delicate but important difference between personalized recommendation and link prediction, we further propose an improved framework named as self-included collaborative filtering (SCF). The SCF framework significantly improved the accuracy and robustness of well-known local indices. The combination of SCF framework and a simple local index can produce an index with competitive performance and much lower complexity compared with elaborately-designed state-of-the-art algorithms., 19 pages, 3 tables, 1 figure

A novel class of carbon nanotube (CNT)-based nanomaterials has been surging since 1991 due to their noticeable mechanical and electrical properties, as well as their good electron transport properties. This is evidence that the development of CNT-reinforced polymer composites could contribute in expanding many areas of use, from energy-related devices to structural components. As a promising material with a wide range of applications, their poor solubility in aqueous and organic solvents has hindered the utilizations of CNTs. The current state of research in CNTs—both single-wall carbon nanotubes (SWCNT) and multiwalled carbon nanotube (MWCNT)-reinforced polymer composites—was reviewed in the context of the presently employed covalent and non-covalent functionalization. As such, this overview intends to provide a critical assessment of a surging class of composite materials and unveil the successful development associated with CNT-incorporated polymer composites. The mechanisms related to the mechanical, thermal, and electrical performance of CNT-reinforced polymer composites is also discussed. It is vital to understand how the addition of CNTs in a polymer composite alters the microstructure at the micro- and nano-scale, as well as how these modifications influence overall structural behavior, not only in its as fabricated form but also its functionalization techniques. The technological superiority gained with CNT addition to polymer composites may be advantageous, but scientific values are here to be critically explored for reliable, sustainable, and structural reliability in different industrial needs.

Though deep learning systems have achieved high accuracy in detecting diseases from medical images, few such systems have been deployed in highly automated disease screening settings due to lack of trust in how well these systems can generalize to out-of-datasets. We propose to use uncertainty estimates of the deep learning system’s prediction to know when to accept or to disregard its prediction. We evaluate the effectiveness of using such estimates in a real-life application for the screening of diabetic retinopathy. We also generate visual explanation of the deep learning system to convey the pixels in the image that influences its decision. Together, these reveal the deep learning system’s competency and limits to the human, and in turn the human can know when to trust the deep learning system.

Ischemic stroke is a leading cause of death and long-term disability that is difficult to predict reliably. Retinal fundus photography has been proposed for stroke risk assessment, due to its non-invasiveness and the similarity between retinal and cerebral microcirculations, with past studies claiming a correlation between venular caliber and stroke risk. However, it may be that other retinal features are more appropriate. In this paper, extensive experiments with deep learning on six retinal datasets are described. Feature isolation involving segmented vascular tree images is applied to establish the effectiveness of vessel caliber and shape alone for stroke classification, and dataset ablation is applied to investigate model generalizability on unseen sources. The results suggest that vessel caliber and shape could be indicative of ischemic stroke, and sourcespecific features could influence model performance.