Browse Articles

Model interpretability on private-safe oriented student dropout prediction

Student dropout is a significant social issue with extensive implications for individuals and society, including reduced employability and economic downturns, which, in turn, drastically influence social sustainable development. Identifying students at high risk of dropping out is a major challenge for sustainable education. While existing machine learning and deep learning models can effectively predict dropout risks, they often rely on real student data, raising ethical concerns and the risk of information leakage. Additionally, the poor interpretability of these models complicates their use in educational management, as it is difficult to justify identifying a student as high-risk based on an opaque model. To address these two issues, we introduced for the first time a modified Preprocessed Kernel Inducing Points data distillation technique (PP-KIPDD), specializing in distilling tabular structured dataset, and innovatively employed the PP-KIPDD to reconstruct new samples that serve as qualified training sets simulating student information distributions, thereby preventing student privacy information leakage, which showed better performance and efficiency compared to traditional data synthesis techniques such as the Conditional Generative Adversarial Networks. Furthermore, we empower the classifiers credibility by enhancing model interpretability utilized SHAP (SHapley Additive exPlanations) values and elucidated the significance of selected features from an educational management perspective. With well-explained features from both quantitative and qualitative aspects, our approach enhances the feasibility and reasonableness of dropout predictions using machine learning techniques. We believe our approach represents a novel end-to-end framework of artificial intelligence application in the field of sustainable education management from the view of decision-makers, as it addresses privacy leakage protection and enhances model credibility for practical management implementations.

Identification and control of booming noise of thin-walled metal plates of commercial vehicle body subjected to high-speed airflow

The thin walled metal plates (TWMP) on top of a commercial vehicle body are easy to produce vibration and booming noise under the action of high-speed airflow, which reduces the sound quality inside the vehicle. Therefore, the effective control of the noise of TWMP is very important for vehicle comfort performance. Based on experimental, finite element, dynamic and fluid techniques, a method for identification, analysis and optimization of TWMP booming noise under high-speed airflow is proposed in this paper. The problem frequency, the noise source and the cause of booming noise inside the vehicle are identified by amplitude-frequency characteristics analysis and modal analysis. By establishing the dynamic model of damping patch, the matching method of damping patch is proposed. The simulation results are in good agreement with the experimental results, which shows that the method is correct. And the damping coefficient, attachment position and attachment area of the damping plate can be analyzed quickly and accurately by using the method, and the vibration and noise of TWMP can be reduced obviously.

Study of heterogeneity in immune responses to exposure in breast cancer: A protocol for a systematic review

Background Radiation is a treatment modality of interest in both palliative and curative settings for patients with breast cancer (BC). Despite many advances in radiation oncology, anti-radiation resistance remains a problem in mechanistic aspects. Studies show that immune cells respond to radiation in a heterogeneous manner and suggest the determinant role of these responses in the ultimate clinical classification of patients with BC as anti-radiation resistant/sensitive. Objective A systematic review of in vivo and in vitro studies will be defined here to assess the radiation response of immune cells in BC comprehensively. Methods/design A systematic search for in vivo and in vitro studies will collect studies that compare the radiation alone-treated and control (untreated or sham) subjects, as well as subjects treated with radiation in combination and subjects treated with radiation alone. The primary outcome of interest will be measures of the abundance of immune cell subsets. The secondary outcomes will include measures derived from co-culture assays leading to immunoregulatory roles or mechanisms proposed for the potential cross-talks between immune-BC cells that will be influenced by radiation. The review results will be primarily synthesized in a narrative and qualitative manner. Meta-analysis will be considered if there are three or more observations with data available for a specific outcome. We will use the SYRCLE “Risk of Bias” assessment tool for appraising the risk of bias in preclinical in vivo studies, and given the lack of standardized instruments for in vitro studies, we will design a checklist suitable for our specific in vitro research questions and publish it along with our systematic review.

Fluorometric combination analysis of conformational changes in urea-induced unfolding of soybean protein isolates and their complexes with casein

Soybean proteins isolates (SPIs) could misfold to form amorphous aggregates after unfolding due to processing conditions, leading to a decrease in their solubility, and casein might be able to block the formation of amorphous aggregates. The turbidity curves showed that SPIs/7S/11S could maintain a soluble state in the presence of casein. The folding and unfolding processes of SPIs are not a single-step event. Fluorescence phase diagram method was utilized to analyze the unfolding process of SPIs and SPI-casein complexes. The results showed that there was one intermediate (molten globule, MG) in the unfolding process of SPIs, while there were two intermediates (dimer or trimer intermediates and MG) in that of SPI-casein complexes. Further attempts were made to explain the conformational changes of SPIs when binding with casein. The surface hydrophobicity index indicated that the intermediate with strong hydrophobicity was exposed when induced by 2 mol/L urea. The fluorescence quenching experiments using KI as a quencher showed that the Ksv of SPIs and 11S and their complexes with casein all reached peak values when urea concentration was 2 mol/L, while Ksv values of 7S and 7S–casein complex peaked at urea concentration 8mol/L and 6mol/L, respectively. Besides, Ksv and fm changing trends of 7S–casein complex was similar to that of SPIs–casein. Based on the above results, it could be inferred that the interactions between SPIs and casein mainly occurs in 7S. The molecular docking of 7S chains from SPIs and casein showed that SPIs could form a “pocket” conformation to “hold” the casein chain. These findings provide new perspectives for understanding the mechanism of interaction between soy proteins and casein, and may provide a basis for research on the functionality of proteins in the food industry. Note: Ksv is the quenching constant of the quenching agent, and fm is the quenching rate.

Effects of preoperative topical levofloxacin on conjunctival microbiome in patients undergoing intravitreal injections

Purpose Endophthalmitis is a serious eye infection that can occur after intravitreal injections. Topical antibiotics are frequently used as a preventative measure, but their impact on the conjunctival microbiome is not fully understood. Methods Conjunctival swabs were collected from 33 eyes of 33 patients undergoing intravitreal injections, both before and after a 3-day course of prophylactic topical levofloxacin 0.5%. Conjunctival microbiome analysis was conducted using 16S rRNA sequencing on the Illumina MiSeq platform. Bioinformatics processing identified unique amplicon sequence variants (ASVs) to evaluate microbial diversity and community composition. Alpha and beta diversity indices were analyzed. Results Topical levofloxacin treatment resulted in no significant change in alpha diversity indices, including Shannon index, Chao1, Shannon, PD whole tree, and observed ASVs, indicating stable microbial richness and evenness. In contrast, beta diversity analysis, assessed through Bray-Curtis dissimilarity, revealed significant differences in microbial composition between pre- and post-treatment samples. These changes included a decrease in the abundance of Staphylococcus and Bacillus species and an increase in the abundance of Streptococcus, Haemophilus, and Neisseria. Conclusion Although prophylactic topical levofloxacin was found to alter the conjunctival microbiome and showed inconsistent effects on the abundance of pathogenic bacteria, its clinical effectiveness as a preventative measure against endophthalmitis remains inconclusive. Further studies are needed to clarify its role in infection prevention.

Virtual reality-based training for radiopharmaceutical administration: development and educational effectiveness

In Japan, task shifting and sharing are promoted to reduce the workload of physicians. Radiological technologists have been assigned new responsibilities, such as securing venous access for radiopharmaceutical administration. This study aimed to develop a prototype Virtual Reality (VR) training system that allows safe and repeatable training for radiological technologists. Additionally, the educational effectiveness of VR training was evaluated, and the concentration levels of the participants were assessed using multiple physiological indicators. Overall, 17 male and 12 female participants (mean age 20.1 ±  0.9 years) were enrolled in this study and classified into two groups—video-based and immersive VR system groups—both of which simulated radiopharmaceutical administration. Concentration and tension levels were evaluated using electroencephalography (EEG) data, salivary amylase levels, and mood assessments. The educational effectiveness was evaluated using a multiple-choice cognitive test. Compared with the resting levels, the alpha/beta ratio of EEG (indicating relaxed concentration) was significantly decreased by 19% in the video-based VR and increased by 40% in the immersive VR groups (both p < 0.05). No significant difference was observed in salivary amylase levels between the two groups. The cognitive test scores, increased by 2.0 and 3.4 points in the video-based VR and immersive VR groups, respectively; a significant difference was observed between both groups (p <  0.05). However, no correlation was found between the EEG ratio and test performance. Thus, immersive VR promotes a more relaxed and concentrated state and was found to have higher educational effectiveness than video-based VR. This suggests that participatory VR training may be more effective than observational VR training. Further research should explore the relationship between educational effectiveness and the evaluation of medical skills.

Complementary Ribo-seq approaches map the translatome and provide a small protein census in the foodborne pathogen Campylobacter jejuni

Abstract In contrast to transcriptome maps, bacterial small protein (≤50-100 aa) coding landscapes, including overlapping genes, are poorly characterized. However, an emerging number of small proteins have crucial roles in bacterial physiology and virulence. Here, we present a Ribo-seq-based high-resolution translatome map for the major foodborne pathogen Campylobacter jejuni. Besides conventional Ribo-seq, we employed translation initiation site (TIS) profiling to map start codons and also developed a translation termination site (TTS) profiling approach, which revealed stop codons not apparent from the reference genome in virulence loci. Our integrated approach combined with independent validation expanded the small proteome by two-fold, including CioY, a new 34 aa component of the CioAB oxidase. Overall, our study generates a high-resolution annotation of the C. jejuni coding landscape, provided in an interactive browser, and showcases a strategy for applying integrated Ribo-seq to other species to enrich our understanding of small proteomes.

Disentangling edge and bulk spin-to-charge interconversion in MoS2 monolayer flakes

Abstract Semiconductor transition metal dichalcogenides are an archetype for spintronic devices due to their spin-to-charge interconversion mechanisms. However, the exact microscopic origin of this interconversion is not yet determined. In our study, we investigated light-induced spin pumping in YIG/MoS 2 heterostructures. Our findings revealed that the MoS 2 monolayer microsized flakes contribute to spin current injection through two distinct mechanisms: metallic edge states and semiconductor area states. The competition between these mechanisms, influenced by the flake size, leads to different behaviors of spin-pumping. Our calculations of the local density of states, by means of density functional theory, of a flake show that light-driven spin current injection can be controlled based on the intensity of light with a suitable wavelength. We demonstrate that a lightdriven spin current injection can enhance up to very high values, attenuate, or even switch on/off the spin-to-charge interconversion. These results hold promise for developing low energy-consuming opto-spintronic device applications.

Alkyne dimerization-hydroarylation to form pentasubstituted 1,3-dienes via binuclear nickel catalysis

Updating the sub-nanometric cognition of reconstructed oxyhydroxide active phase for water oxidation

Cryo-TEM structure of β-glucocerebrosidase in complex with its transporter LIMP-2

Abstract Targeting proteins to their final cellular destination requires transport mechanisms and nearly all lysosomal enzymes reach the lysosome via the mannose-6-phosphate receptor pathway. One of the few known exceptions is the enzyme β-glucocerebrosidase (GCase) that requires the lysosomal integral membrane protein type-2 (LIMP-2) as a proprietary lysosomal transporter. Genetic variations in the GCase encoding gene GBA1 cause Gaucher’s disease (GD) and present the highest genetic risk factor to develop Parkinson’s disease (PD). Activators targeting GCase emerge as a promising therapeutic approach to treat GD and PD, with pre-clinical and clinical trials ongoing. In this study, we resolve the complex of GCase and LIMP-2 using cryo-electron microscopy with the aid of an engineered LIMP-2 shuttle and two GCase-targeted pro-macrobodies. We identify helix 5 and helix 7 of LIMP-2 to interact with a binding pocket in GCase, forming a mostly hydrophobic interaction interface supported by one essential salt bridge. Understanding the interplay of GCase and LIMP-2 on a structural level is crucial to identify potential activation sites and conceptualizing novel therapeutic approaches targeting GCase. Here, we unveil the protein structure of a mannose-6-phosphate-independent lysosomal transport complex and provide fundamental knowledge for translational clinical research to overcome GD and PD.

Quantitative PET imaging and modeling of molecular blood-brain barrier permeability

Abstract Neuroimaging of blood-brain barrier permeability has been instrumental in identifying its broad involvement in neurological and systemic diseases. However, current methods evaluate the blood-brain barrier mainly as a structural barrier. Here we developed a non-invasive positron emission tomography method in humans to measure the blood-brain barrier permeability of molecular radiotracers that cross the blood-brain barrier through its molecule-specific transport mechanism. Our method uses high-temporal resolution dynamic imaging and kinetic modeling for multiparametric imaging and quantification of the blood-brain barrier permeability-surface area product of molecular radiotracers. We show, in humans, our method can resolve blood-brain barrier permeability across three radiotracers and demonstrate its utility in studying brain aging and brain-body interactions in metabolic dysfunction-associated steatotic liver inflammation. Our method opens new directions to effectively study the molecular permeability of the human blood-brain barrier in vivo using the large catalogue of available molecular positron emission tomography tracers.

Single-cell sequencing uncovers a high ESM1-expression endothelial cell subpopulation associated with bladder cancer progression and the immunosuppressive microenvironment

Impact of optimized and conventional facility designs on outpatient abdominal MRI workflow efficiency

Effect of SiO2 aerogel and perlite on properties of foamed concrete

SGA-Driven feature selection and random forest classification for enhanced breast cancer diagnosis: A comparative study

Multimodal contrastive learning for enhanced explainability in pediatric brain tumor molecular diagnosis

Growth variability of farm grown teak in response to climatic and soil factors across three agroclimatic zones of Tamil Nadu, India

Ultra-lightweight robotic hip exoskeleton with anti-phase torque symmetry for enhanced walking efficiency

Abstract This paper presents a novel robotic exoskeleton that is exceptionally lightweight and compact, while providing effective gait assistance. To maximize the system’s assistance-to-weight/size ratio, the design focuses on two key aspects of human gait mechanics: (1) the contribution of the hip joints to power generation, and (2) the symmetrical nature of hip flexion and extension torques during walking. Based on these principles, we developed a compact hip exoskeleton with a single actuator. This actuator simultaneously drives the hip joint in the sagittal plane, facilitating both flexion and extension during gait. An adaptive delayed output feedback controller was implemented, ensuring consistent performance across diverse walking conditions using a single rotational sensor and actuator. To evaluate the exoskeleton’s effectiveness, a 4-week outdoor walking exercise program was conducted with nine elderly participants. Their gait, balance, and muscle strength were measured before and after the program to assess improvements. Results showed significant improvements in walking speed (14.8% in the 10-m walk and 10.6% in the 6-min walk), as well as enhanced performance in the timed up-and-go test (24.5%) and the short physical performance battery test (18.7%). Ankle dorsiflexion and plantar flexion muscle strength increased by 75.45% and 45.8%, respectively. Additionally, metabolic measurements from three young adults indicated a 13.6 ± 3.2% reduction in the net metabolic cost of walking with the exoskeleton compared to walking without it. These results demonstrate that the single actuator-based hip exoskeleton offers effective gait assistance while maintaining a lightweight and compact design, highlighting its potential for widespread use in various applications.

Effect of nanosecond laser assisted surface modification on physical and mechanical properties of denture base materials

Abstract Lasers are being used for modifying the surfaces of biomaterials to make them resistant to microbial adhesion. However, the effect of such surface modification on physical and mechanical properties has not been widely reported. The aim of the study was to investigate the effect of surface modification using laser patterning on the physical and mechanical properties of denture base materials. Nd: YAG nanosecond laser was used to create different patterns on two commercial denture base materials. Surface characteristics of the patterned specimens such as surface roughness, contact angle and resistance to microbial adhesion were measured. Laser patterned specimens with lower microbial adhesion were subjected to evaluation of mechanical (flexural strength and surface hardness) and physical (water sorption and solubility) properties using standard methods. Laser patterning increased the surface roughness, contact angle and the resistance to microbial adhesion. Laser patterning did not have any detrimental effects on mechanical and physical properties. However, a significant increase in surface hardness was observed in all patterned specimens. By fine-tuning the laser patterning parameters, it is possible to create surfaces with enhanced resistance to microbial adhesion without compromising the physical and mechanical properties of the material, which can ultimately lead to better oral health outcomes for denture users.