Boosting-Crystal Graph Convolutional Neural Network for Predicting Highly Imbalanced Data: A Case Study for Metal-Insulator Transition Materials.

Applying machine-learning techniques for imbalanced data sets presents a significant challenge in materials science since the underrepresented characteristics of minority classes are often buried by the abundance of unrelated characteristics in majority of classes. Existing approaches to address this focus on balancing the counts of each class using oversampling or synthetic data generation techniques. However, these methods can lead to loss of valuable information or overfitting. Here, we introduce a deep learning framework to predict minority-class materials, specifically within the realm of metal-insulator transition (MIT) materials. The proposed approach, termed boosting-CGCNN, combines the crystal graph convolutional neural network (CGCNN) model with a gradient-boosting algorithm. The model effectively handled extreme class imbalances in MIT material data by sequentially building a deeper neural network. The comparative evaluations demonstrated the superior performance of the proposed model compared to other approaches. Our approach is a promising solution for handling imbalanced data sets in materials science.

Crystallographic Pathways to Tailoring Metal-Insulator Transition through Oxygen Transport in VO2.

The metal-insulator (MI) transition of vanadium dioxide (VO2) is effectively modulated by oxygen vacancies, which decrease the transition temperature and insulating resistance. Oxygen vacancies in thin films can be driven by oxygen transport using electrochemical potential. This study delves into the role of crystallographic channels in VO2 in facilitating oxygen transport and the subsequent tuning of electrical properties. A model system is designed with two types of VO2 thin films: (100)- and (001)-oriented, where channels align parallel and perpendicular to the surface, respectively. Growing an oxygen-deficient TiO2 layer on these VO2 films prompted oxygen transport from VO2 to TiO2. Notably, in (001)-VO2 film, where oxygen ions move along the open channels, the oxygen migration deepens the depleted region beyond that in (100)-VO2, leading to more pronounced changes in metal-insulator transition behaviors. The findings emphasize the importance of understanding the intrinsic crystal structure, such as channel pathways, in controlling ionic defects and customizing electrical properties for applications.

Elucidating the Role of Alkali Metal Carbonates in Impact on Oxygen Vacancies for Efficient and Stable Perovskite Solar Cells.

Effectively suppressing nonradiative recombination at the SnO2/perovskite interface is imperative for perovskite solar cells. Although the capabilities of alkali salts at the SnO2/perovskite interface have been acknowledged, the effects and optimal selection of alkali metal cations remain poorly understood. Herein, a novel approach for obtaining the optimal alkali metal cation (A-cation) at the interface is investigated by comparatively analyzing different alkali carbonates (A2CO3; Li2CO3, Na2CO3, K2CO3, Rb2CO3, and Cs2CO3). Theoretical calculations demonstrate that A2CO3 coordinates with undercoordinated Sn and O on the surface, effectively mitigating oxygen vacancy (VO) defects with increasing A-cation size, whereas Cs2CO3 exhibits diminished preferability owing to enhanced steric hindrance. The experimental results highlight the crucial role of Rb2CO3 in actively passivating VO defects, forming a robust bond with SnO2, and facilitating Rb+ diffusion into the perovskite layer, thereby enhancing charge extraction, alleviating deep-level trap states and structural distortion in the perovskite film, and significantly suppressing nonradiative recombination. X-ray absorption spectroscopy analyses further reveal the effect of Rb2CO3 on the local structure of the perovskite film. Consequently, a Rb2CO3-treated device with aperture area of 0.14 cm2 achieves a notable efficiency of 22.10%, showing improved stability compared to the 20.11% achieved for the control device.

Glycemic Control and Oral Health Outcomes in Patients With Diabetes: Insights From a Nationwide Korean Survey.

BACKGROUND: Diabetes is recognized as a risk factor for various inflammatory conditions, including periodontitis. There exists a bidirectional relationship between glycemic control and oral health in individuals with diabetes. This study aimed to analyze the link between glycemic control and oral health status among Korean patients with diabetes. METHODS: Using data from a population-based nationwide survey conducted between 2007 and 2019, we identified 70,554 adults with diabetes-related information. The study population included 9,090 individuals diagnosed with diabetes and 61,164 healthy controls. The association between glycemic control, defined by mean glycated hemoglobin (HbA1c) values, and various oral health measures, such as tooth brushing frequency, periodontitis, denture wearing, Decayed, Missing, and Filled Teeth (DMFT) index, number of remaining teeth, and past-year dental clinic visits, was evaluated using multivariate logistic regression analyses. RESULTS: Compared to the control group, patients with diabetes exhibited a higher prevalence of periodontitis (88.6% vs. 73.3%), complete dentures (5.0% vs. 1.5%), and elevated DMFT index (33.2% vs. 26.7%) (all P < 0.001). Multivariate analyses revealed significant associations between diabetes and several oral health factors: denture status (No denture: adjusted odds ratio [aOR], 0.784; 95% confidence interval [CI], 0.627-0.979), and having fewer permanent teeth (0-19) (aOR, 1.474; 95% CI, 1.085-2.003). Additionally, a positive correlation was found between higher HbA1c levels and the risk of having fewer remaining teeth (0-19) (HbA1c < 6.5%: aOR, 1.129; 95% CI, 0.766-1.663; 6.5% ≤ HbA1c < 8.0%: aOR, 1.590; 95% CI, 1.117-2.262; HbA1c ≥ 8%: aOR, 1.910; 95% CI, 1.145-3.186) (P for trends = 0.041). CONCLUSION: We found a positive association between diabetes and poor oral health, as well as a noteworthy relationship between reduced permanent teeth (≤ 19) and glycemic control. These insights emphasize the critical role of oral health management in diabetic care and underscore the importance of maintaining effective glycemic control strategies for overall health and well-being in patients with diabetes.

The Efficacy and Safety of Sodium-Glucose Co-transporter 2 (SGLT2) Inhibitors in Real-World Clinical Practice: Potential Cautionary Use in Elderly Patients with Type 2 Diabetes (T2D).

INTRODUCTION: Sodium-glucose co-transporter 2 (SGLT2) inhibitors have shown safe and therapeutic efficacy in randomized controlled trials (RCT) to reduce adverse cardiorenal events in high-risk patients with type 2 diabetes (T2D). In this study, we investigated the efficacy and safety of SGLT2 intervention in patients with T2D in a real-world clinical practice to confirm the validity of the RCT results. METHODS: As a retrospective study, we evaluated medical records from 596 patients with T2D treated with SGLT2 inhibitors (dapagliflozin or empagliflozin) in addition to their prior drug regimen to improve glucose control between 2015 and 2019 in the Endocrinology Department at Chungbuk National University Hospital. No control arm was evaluated to compare the effects of adding SGLT inhibitors to the pre-existing regimen. The primary objective was the measurement of glycated hemoglobin (HbA1c) from each individual patient over a 36-month period at 6-month intervals. The secondary parameters were the measurement of fasting plasma glucose (FPG) and body weight (Bwt) changes, as well as the monitoring of adverse events (AEs) and determining the reasons for drug discontinuation. RESULTS: HbA1c levels were reduced at each of the time points throughout the 36-month period and were significantly reduced by 12.5% (P < 0.01) from time 0 (8.8 ± 1.3%) to 36 months (7.7 ± 1.0%). FPG levels [from basal (180 ± 60 mg/dL) to 36 months (138 ± 38 mg/dL)] and Bwt [from basal (74 ± 15 kg) to 36 months (72 ± 15 kg)] were also significantly reduced (P < 0.01) for both measurements in the SGLT2 inhibitor add-on group. Similar to HbA1c profile, the FPG and Bwt were measured at a consistently lower level at 6 months until the end of the study. The most common AEs were hypoglycemia (n = 57), genitourinary infection (GUI) (n = 31), and polyuria (n = 28). In the elderly population (≥ 75 years old), AEs (31%) were generally more prevalent (P < 0.001) than those (21%) in the adult (< 75 years old) patients. Over the study period, 211 (35%) patients either dropped or completely discontinued the use of the SGLT2 inhibitor, and the elderly patients tended to have a higher discontinuation rate (52%; P = 0.005) than the adults (33%). CONCLUSIONS: In this study, we demonstrated that SGLT2 inhibitors are an effective and durable hypoglycemic agent to control blood glucose levels with reduced maintenance of Bwt, but their use in the elderly (≥ 75 years old) patients with T2D may warrant some additional caution due to increased probability of AEs and discontinuation of drug use.

Unveiling the Role of Side Chain for Improving Nonvolatile Characteristics of Conjugated Polymers-Based Artificial Synapse.

Interest has grown in services that consume a significant amount of energy, such as large language models (LLMs), and research is being conducted worldwide on synaptic devices for neuromorphic hardware. However, various complex processes are problematic for the implementation of synaptic properties. Here, synaptic characteristics are implemented through a novel method, namely side chain control of conjugated polymers. The developed devices exhibit the characteristics of the biological brain, especially spike-timing-dependent plasticity (STDP), high-pass filtering, and long-term potentiation/depression (LTP/D). Moreover, the fabricated synaptic devices show enhanced nonvolatile characteristics, such as long retention time (≈102 s), high ratio of Gmax/Gmin, high linearity, and reliable cyclic endurance (≈103 pulses). This study presents a new pathway for next-generation neuromorphic computing by modulating conjugated polymers with side chain control, thereby achieving high-performance synaptic properties.

Energy-Efficient and Self-Powered Green Ammonia Synthesis by Electrochemical Nitrate Reduction Combined with Hydrazine Oxidation.

To achieve the global goal of carbon neutrality, recently, emphasis has been placed on developing green ammonia production method to replace the Haber-Bosch process. Nitrate reduction reaction (NO3 RR) has received considerable attention, especially for electrochemically producing ammonia from nitrate and simultaneously purifying wastewater. This study first demonstrates that the combination of NO3 RR with hydrazine oxidation reaction (HzOR) is an energy efficient green ammonia production method, which overcomes the sluggish water oxidation limitation. Tungsten phosphide (WP) nanowires (NWs) are prepared as cathode NO3 RR electrocatalysts, which exhibit a high Faradaic efficiency in both neutral (≈93%) and alkaline (≈85%) media. Furthermore, they show a high bifunctional activity in anodic reactions and exhibit a low potential 0.024 V for generating a current density of 10 mA cm-2 in HzOR. The overall NO3 RR-HzOR required an impressively low potential of 0.24 V for generating a current density of 10 mA cm-2 ; this potential is much lower than those required for NO3 RR-OER (1.53 V) and NO3 RR-UOR (1.31 V). A self-powered ammonia production system, prepared by assembling an NO3 RR-HzOR with a perovskite solar cell, displays a high ammonia production rate of 1.44 mg cm-2  h-1 . A single PV cell provides enough driving voltage in the PV-EC due to low required potential. This system facilitates unassisted green ammonia synthesis with a low energy consumption and also allows upcycling of wastewater to produce useful fuel.

First-Principles Investigations on the Semiconductor-to-Metal Phase Transition of 2D Si2Te3 for Reversible Resistive Switching.

Si2Te3 is attracting attention due to its compatibility with Si technology while still showing advantages as a two-dimensional layered material. Although recent experimental studies have observed the resistive switching process in Si2Te3-based memristors, the mechanism has not been clearly identified. In this study, first-principles density functional theory calculations are employed to understand the relationship between the phase transition of Si2Te3 and the reversible resistive switching of the Si2Te3-based memristor. Our calculation results show that although semiconducting Si2Te3 is energetically more stable than two metallic Si2Te3 phases (α and ß), two metallic Si2Te3 can be energetically stabilized by excess holes. The enhanced energetic preference of two metallic Si2Te3 by excess holes is explained by the reduced occupation of antibonding states between Si and Te. Our study finds that the energy barrier for the phase transition between semiconducting Si2Te3 and α-metallic Si2Te3 varies significantly by excess charge carriers so the phase transition can be directly connected to the reversible resistive switching of the Si2Te3-based memristor under external bias. Our finding will serve as a cornerstone for optimizing the resistive switching process of the Si2Te3-based memristor.

Influenza vaccination trend and related factors among patients with diabetes in Korea: Analysis using a nationwide database.

Background: Subjects with diabetes are at higher risk of serious influenza-related complications. We aimed to investigate the yearly trend of influenza vaccination and factors associated with being unvaccinated for influenza in subjects with diabetes using a nationwide observational study performed within the recent decade. Methods: Among 105,732 subjects from the Korea National Health and Nutrition Examination Survey between 2007 and 2019, 8,632 with diabetes were included. We investigated the yearly trend of influenza vaccination and factors associated with being unvaccinated for influenza. Results: During the study period, the prevalence of influenza vaccination in subjects with diabetes showed a tendency to increase every year, reaching almost 60% in 2019, which was higher than the rate in subjects without diabetes. Younger age (adjusted hazard ratio (aHR) [95% CI] 11.29 [8.63-14.75] for < 50 years; 6.16 [5.21-7.29] for 50-65 years), male (aHR 1.67 [1.52-1.87]), current smoker (aHR 1.31 [1.00-1.72], lower-income status (aHR 1.46 [1.17, 1.84]), and high education level (aHR 1.30 [1.01-1.67]) were associated with being unvaccinated. Also, a poorer glycemic control with HbA1c ≥ 9% was found to be correlated with unvaccinated status (aHR 1.48 [1.15-1.90]). Conclusion: The influenza vaccination rate is still unsatisfactory in subjects with diabetes. Young age, males, low-income level, high education level, and poor glycemic control were associated with unvaccinated status. Considering the risk-benefits of influenza vaccination in patients with diabetes, physicians should make an effort to increase vaccination rates, especially in low vaccination rate groups.

Drastic Gas Sensing Selectivity in 2-Dimensional MoS2 Nanoflakes by Noble Metal Decoration.

Noble metal nanoparticle decoration is a representative strategy to enhance selectivity for fabricating chemical sensor arrays based on the 2-dimensional (2D) semiconductor material, represented by molybdenum disulfide (MoS2). However, the mechanism of selectivity tuning by noble metal decoration on 2D materials has not been fully elucidated. Here, we successfully decorated noble metal nanoparticles on MoS2 flakes by the solution process without using reducing agents. The MoS2 flakes showed drastic selectivity changes after surface decoration and distinguished ammonia, hydrogen, and ethanol gases clearly, which were not observed in general 3D metal oxide nanostructures. The role of noble metal nanoparticle decoration on the selectivity change is investigated by first-principles density functional theory (DFT) calculations. While the H2 sensitivity shows a similar tendency with the calculated binding energy, that of NH3 is strongly related to the binding site deactivation due to preferred noble metal particle decoration at the MoS2 edge. This finding is a specific phenomenon which originates from the distinguished structure of the 2D material, with highly active edge sites. We believe that our study will provide the fundamental comprehension for the strategy to devise the highly efficient sensor array based on 2D materials.

Development research on an AI English learning support system to facilitate learner-generated-context-based learning.

For decades, AI applications in education (AIEd) have shown how AI can contribute to education. However, a challenge remains: how AIEd, guided by educational knowledge, can be made to meet specific needs in education, specifically in supporting learners' autonomous learning. To address this challenge, we demonstrate the process of developing an AI-applied system that can assist learners in studying autonomously. Guided by a Learner-Generated Context (LGC) framework and development research methodology (Richey and Klein in J Comput High Educ 16(2):23-38, https://doi.org/10.1007/BF02961473, 2005), we define a form of learning called "LGC-based learning," setting specific study objectives in the design, development, and testing of an AI-based system that can facilitate Korean students' LGC-based English language learning experience. The new system is developed based on three design principles derived from the literature review. We then recruit three Korean secondary-school students with different educational backgrounds and illustrate and analyze their English learning experiences using the system. Following this analysis, we discuss how the AI-based system facilitates LGC-based learning and further issues to be considered for future research.

Correction: Optical humidity sensors based on lead-free Cu-based perovskite nanomaterials.

[This corrects the article DOI: 10.1039/D2NA00168C.].

Anionic Flow Valve Across Oxide Heterointerfaces by Remote Electron Doping.

As an analogue of charged electron flows, the ionic flow could be controlled by the electronic band alignment due to the ambipolar nature of diffusion in the ionic crystal. Here, we demonstrate the active control of the anionic diffusion across heterointerfaces through remote electron doping in the capping layers. In contrast to the spontaneous ionic flux from the underlying VO2 layers to the undoped TiO2 capping layers, the activated Nb dopants in the TiO2 capping layers substantially restrict the ionic flux, despite identical growth conditions. The increase of Fermi level by Nb donors in TiO2 prevents electron flux from being generated across the interfaces by the heightening of a Schottky barrier; this electron shortage generates a kinetic close valve for the flow of negatively charged oxygen ions. Thus, these results demonstrate the importance of electron supply on charged ionic flow, thereby suggesting an unprecedented strategy for ionic-defect-induced emergent properties at interfaces.

Association between the triglyceride-glucose (TyG) index and increased blood pressure in normotensive subjects: a population-based study.

BACKGROUND: Insulin resistance (IR) is an important contributor to the development of hypertension (HTN), and the triglyceride-glucose (TyG) index has been proposed as a simple, reliable marker of IR. This study investigated the association between the TyG index and blood pressure (BP) elevation in a large general population. METHODS: The study enrolled 15,721 adults with no history of cardiometabolic diseases from the 2016-2019 Korea National Health and Nutrition Examination Survey. Participants were classified into quartiles based on the TyG index and BP was categorized as normal BP, elevated BP, pre-HTN, and HTN. The associations of the TyG index with BP categories were assessed using multivariate multinomial logistic regression models with normal BP as the reference group. RESULTS: The mean systolic/diastolic BP and prevalence of HTN increased with the TyG index (P for trend < 0.001). The continuous TyG index had a strong dose-response relationship with increased odds of elevated BP, pre-HTN, and HTN. Compared with the lowest TyG index quartile, the highest TyG index quartile was significantly associated with higher odds of having elevated BP (odds ratio [OR], 1.52; 95% confidence interval [CI], 1.24-1.87; P for trend < 0.001), pre-HTN (OR, 2.22; 95% CI, 1.95-2.53; P for trend < 0.001), and HTN (OR, 4.24; 95% CI, 3.49-5.16; P for trend < 0.001). CONCLUSION: We found that a higher TyG index was positively associated with the risk of increased BP in normal healthy individuals. This study suggests that the TyG index might serve as a potential predictor of HTN. However, further studies with larger sample sizes and various target populations in longitudinal designs are needed.

Noncovalent minimal assembly of exogenous histamine with hemin cofactor as a peroxidase-mimicking cooperative catalyst.

By mimicking the synergistic interplay of primary and secondary coordination spheres within native peroxidases, we demonstrate a scaffold-free, yet highly effective molecular-level cooperation between an iron(III)-containing hemin cofactor and exogenous histamine in accelerating a peroxidase-like reaction. Density functional theory computations predict that, among structurally similar molecules, the histamine is the most interactive partner of hemin to elicit a spontaneous peroxidation by electrostatically attracting the proton of hydrogen peroxide to its own imidazole and thermodynamically stabilizing a transition-state intermediate. Although the molecular weight of hemin-histamine pair is 763, 1.7% of the horseradish peroxidase, cooperative catalysis of two natural molecules exhibits 17.3 times greater catalytic efficiency (17.93 M-1s-1) and 57.8 times larger specific activity (36.45 µmol/min·mg) than the hemin alone (1.04 M-1s-1 and 0.63 µmol/min·mg). Despite no scaffold or covalent linkage, the self-assembly with hemin is highly histamine-specific in complex environments, leading rapid color changes by substrate oxidation within 10 s.

Entangling three identical particles via spatial overlap.

Quantum correlations between identical particles are at the heart of quantum technologies. Several studies with two identical particles have shown that the spatial overlap and indistinguishability between the particles are necessary for generating bipartite entanglement. On the other hand, researches on the extension to more than two-particle systems are limited by the practical difficulty to control multiple identical particles in laboratories. In this work, we propose schemes to generate two fundamental classes of genuine tripartite entanglement, i.e., GHZ and W classes, which are experimentally demonstrated using linear optics with three identical photons. We also show that the tripartite entanglement class decays from the genuine entanglement to the full separability as the particles become more distinguishable from each other. Our results support the prediction that particle indistinguishability is a fundamental element for entangling identical particles.

Optical humidity sensors based on lead-free Cu-based perovskite nanomaterials.

Organometallic halide perovskite materials possess unique and tunable optical properties with a wide range of optoelectronic applications. However, these materials suffer from humidity-driven degradation in ambient atmospheres. In this paper we investigate stable copper-based perovskite nanocrystals for potential use in humidity sensors, specifically examining their unique humidity-dependent optical properties and reversibility. We controlled stoichiometric ratios of Cu-based perovskites and demonstrated that (methylammonium)2CuBr4 nanocrystals showed excellent reversible physisorption of water molecules. These perovskite nanocrystals exhibited reversible hydro-optical properties, including transparency changes in response to variations in relative humidity under ambient conditions. The perovskite nanomaterial humidity sensor was highly reliable and stable, with a linear correlation in a relative humidity range of 7% to 98%. Accordingly, the lead-free Cu-based perovskite materials developed herein have the potential to be employed as real-time, self-consistent humidity sensors.

Which factors are helpful for the early determination of treatment level in patients with interstitial lung disease in the intensive care unit to minimize the suffering in their end of life?: A retrospective study.

Interstitial lung disease (ILD) is widely known to be associated with high mortality and poor prognosis, especially in patients admitted to the intensive care unit (ICU). The objective of this study was to investigate clinical predictors for assisting relatively early decision of treatment level in the ICU. We retrospectively investigated patients with ILD who were admitted to the ICU between January 1, 2014, and September 30, 2019. A total of 64 patients were analyzed. We found the ICU and hospital mortality rates to be 67.2% and 69.8%, respectively. Nonsurvivors had a higher fraction of inspired oxygen (FiO2) on days 1 (79 ±â€…21 vs 60% ±â€…21%, P = .001) and 3 (61 ±â€…31 vs 46% ±â€…19%, P = .004). They showed lower partial pressure of oxygen/FiO2 (PF) ratio on days 1 (134 ±â€…80 vs 173 ±â€…102, P = .049) and 3 (147 ±â€…74 vs 235 ±â€…124, P = .003) than the survivor group. The lactic acid levels obtained on day 1 and PF ratio measured on day 3 were associated with mortality (odds ratio, 1.89; 95% confidence interval 1.03-3.47 and odds ratio, 0.99; 95% confidence interval 0.98-1.00, respectively). Among the 31 ICU survivors, 10 patients died in the general ward, 12 patients died after hospital discharge; only 9 patients survived after 1 year. We suggest that these clinical predictors could be used to determine the level of further treatment or withdrawal on day 3 of admission in patients with ILD admitted to the ICU to minimize the prolonged suffering in a relatively early period.

Optimal blood pressure target in the elderly: rationale and design of the HOW to Optimize eLDerly systolic Blood Pressure (HOWOLD-BP) trial.

BACKGROUND/AIMS: The optimal systolic blood pressure (SBP) goal for elderly patients with hypertension, especially to reduce cardiovascular disease (CVD) incidence and improve outcome, is unclear. This study aimed to evaluate the beneficial effects of intensive treatment for hypertension on the incidence of CVD in elderly Korean patients. METHODS: The HOW to Optimize eLDerly systolic Blood Pressure (HOWOLD-BP) trial is a multicenter, parallel-design, open-label, randomized controlled trial designed to evaluate whether intensive treatment (SBP ≤ 130 mmHg) will provide more benefits in lowering the incidence and mortality associated with CVD than standard treatment (SBP ≤ 140 mmHg) in elderly patients with hypertension aged ≥ 65 years. For this study, eleven university hospitals in Korea will enroll approximately 3,176 elderly patients with hypertension between 2019 and 2022. Patients will be requested to visit the clinic every 4 months for the first year and every 6 months thereafter for 36 months. Parameters, including clinic and home blood pressure, anthropometric and laboratory findings, and frailty assessments, will be collected according to the standardized protocol. The primary outcome is a composite of CVD (acute coronary syndrome, stroke, and heart failure) incidence and cardiovascular deaths. RESULTS: As of December 2021, 1,655 participants had been enrolled in the study, including 831 patients in the standard group and 824 patients in the intensive group. CONCLUSION: The HOWOLD-BP trial is the first study performed in Korea to evaluate the beneficial effects of intensive blood pressure treatment on CVD in elderly patients with hypertension. The results of this study will help clarify the appropriate target SBP for this population.