Search

Academic performance prediction aims to leverage student-related information to predict their future academic outcomes, which is beneficial to numerous educational applications, such as personalized teaching and academic early warning. In this article, we reveal the students' behavior trajectories by mining campus smartcard records, and capture the characteristics inherent in trajectories for academic performance prediction. Particularly, we carefully design a tri-branch convolutional neural network (CNN) architecture, which is equipped with rowwise, columnwise, and depthwise convolutions and attention operations, to effectively capture the persistence, regularity, and temporal distribution of student behavior in an end-to-end manner, respectively. However, different from existing works mainly targeting at improving the prediction performance for the whole students, we propose to cast academic performance prediction as a top-k ranking problem, and introduce a top-k focused loss to ensure the accuracy of identifying academically at-risk students. Extensive experiments were carried out on a large-scale real-world dataset, and we show that our approach substantially outperforms recently proposed methods for academic performance prediction. For the sake of reproducibility, our codes have been released at https://github.com/ZongJ1111/Academic-Performance-Prediction.

A small-sized chromophore, BTTA-2OH, manifesting favorable solubility, large two-photon excitation efficiency, and good fluorescence photostability was synthesized to label the membrane of living cells for visualizing the dynamic movement of membrane-related vesicles via a two-photon fluorescence imaging technique based on wavelength-tunable temporal-focusing multiphoton excitation microscopy.

Formation of a large igneous province, an explosive magmatic event in earth, is intimately related to geodynamic processes. The Tarim large igneous province, a famous representative in China, conta...

Objectives To explore the effects of acupoint application therapy (AAT) with TianGui Powder (TGP) on the expressions of the transforming growth factor β1 (TGF‐β1) and Smad‐2/3 signaling pathway in ovariectomized osteoporosis rats. Methods Sixty rats were randomly divided into four groups: normal group (group A), model group (group B), TGP group (group C), and Western medicine group (group D). Group A had only the corresponding amount of adipose tissue around the ovary removed; rats in the other groups had bilateral ovariectomies. After 1 week, groups A and B were given 1 mL/100 mg normal saline solution by gavage, group C was treated with AAT with TGP on ShenQue acupoint (0.2 piece/rat, 6 h/time, 1 time/d) and group D was given calcium carbonate vitamin D3 (36 mg/kg/d) and alfacalcidol (0.05 μg/kg/d) tablet suspension. In this study, the bone mineral density (BMD) , the levels of BALP, TRAP‐5b, and BGP in serum and the changes in bone histomorphology was detected. For acquiring lumbar experimental data, the expression of TGF‐β1, Smad‐2/3 proteins and mRNA of TGF‐β1and Smad‐2/3 were assessed. After 12 weeks, the data were collected for analysis. Results Compared with group A, the bone trabecula was thinner and significantly reduced in other groups. The result of BMD improved significantly in both groups C and D compared to group B after intervention (P < 0.05). In contrast, compared to group B, the levels of BALP, TRAP‐5b, and BGP significantly declined in both groups C and D. In group C, the results of protein expressions in TGF‐β1, Smad‐2/3 were 2.870 ± 0.270, 1.552 ± 0.111, and 1.420 ± 0.079, respectively. In groups C and D, those indications significantly declined compared to group B (P < 0.01). In group C, the level of mRNA expressions of TGF‐β1, Smad‐2/3 were 1.872 ± 0.177, 1.672 ± 0.086, and 1.790 ± 0.136, respectively. Compared with group B, those indications had significant difference in groups C and D (P < 0.05). Conclusion Acupoint application therapy with TGP could significantly improve the BMD. The TGF‐β1 and Smad‐2/3 signaling pathway could be a therapeutic target of TGP in postmenopausal osteoporosis rats.

OBJECTIVES To explore the effects of acupoint application therapy (AAT) with TianGui Powder (TGP) on the expressions of the transforming growth factor β1 (TGF-β1) and Smad-2/3 signaling pathway in ovariectomized osteoporosis rats. METHODS Sixty rats were randomly divided into four groups: normal group (group A), model group (group B), TGP group (group C), and Western medicine group (group D). Group A had only the corresponding amount of adipose tissue around the ovary removed; rats in the other groups had bilateral ovariectomies. After 1 week, groups A and B were given 1 mL/100 mg normal saline solution by gavage, group C was treated with AAT with TGP on ShenQue acupoint (0.2 piece/rat, 6 h/time, 1 time/d) and group D was given calcium carbonate vitamin D3 (36 mg/kg/d) and alfacalcidol (0.05 μg/kg/d) tablet suspension. In this study, the bone mineral density (BMD) , the levels of BALP, TRAP-5b, and BGP in serum and the changes in bone histomorphology was detected. For acquiring lumbar experimental data, the expression of TGF-β1, Smad-2/3 proteins and mRNA of TGF-β1and Smad-2/3 were assessed. After 12 weeks, the data were collected for analysis. RESULTS Compared with group A, the bone trabecula was thinner and significantly reduced in other groups. The result of BMD improved significantly in both groups C and D compared to group B after intervention (P < 0.05). In contrast, compared to group B, the levels of BALP, TRAP-5b, and BGP significantly declined in both groups C and D. In group C, the results of protein expressions in TGF-β1, Smad-2/3 were 2.870 ± 0.270, 1.552 ± 0.111, and 1.420 ± 0.079, respectively. In groups C and D, those indications significantly declined compared to group B (P < 0.01). In group C, the level of mRNA expressions of TGF-β1, Smad-2/3 were 1.872 ± 0.177, 1.672 ± 0.086, and 1.790 ± 0.136, respectively. Compared with group B, those indications had significant difference in groups C and D (P < 0.05). CONCLUSION Acupoint application therapy with TGP could significantly improve the BMD. The TGF-β1 and Smad-2/3 signaling pathway could be a therapeutic target of TGP in postmenopausal osteoporosis rats.

Convolutional Neural Network (CNN) based multi-task learning methods have been widely used in a variety of applications of computer vision. Towards effective multi-task CNN architectures, recent studies automatically learn the optimal combinations of task-specific features at single network layers. However, they generally construct an unchanged operation of feature aggregation after training, regardless of the characteristics of input features. In this paper, we propose a novel Adaptive Feature Aggregation (AFA) layer for multi-task CNNs, in which a dynamic aggregation mechanism is designed to allow each task to adaptively determine the degree to which the feature aggregation of different tasks is needed according to the feature dependencies. On both pixel-level and image-level tasks, we demonstrate that our approach significantly outperforms the previous state-of-the-art methods of multi-task CNNs.

Student performance prediction aims to leverage student-related information to predict their future academic outcomes, which may be beneficial to numerous educational applications, such as personalized teaching and academic early warning. In this paper, we seek to address the problem by analyzing students' daily studying and living behavior, which is comprehensively recorded via campus smart cards. Different from previous studies, we propose an end-to-end student performance prediction model, namely Tri-branch CNN, which is equipped with three types of convolutional filters, i.e., the row-wise convolution, column-wise convolution, and group-wise convolution, to effectively capture the duration, periodicity, and location-aware characteristic of student behavior, respectively. We also introduce the attention mechanism and cost-sensitive learning strategy to further improve the accuracy of our approach. Extensive experiments on a large-scale real-world dataset demonstrate the potential of our approach for student performance prediction.

The purpose of this study is to investigate chemotherapy-related cognitive impairment (CRCI) in breast cancer patients, analyze absolute concentration and structural changes of metabolites in different brain regions by multimodal neuroimaging technology, and explore correlation between them. Breast cancer patients with chemotherapy treatment group (Ctx+, N = 24) and control group without treatment (Ctx-, N = 20) underwent neuropsychological tests, multivoxel magnetic resonance spectroscopy (MRS), and diffusion tensor imaging (DTI) before and after chemotherapy. Regions of interest (ROls) in magnetic resonance scan include bilateral posterior cingulate gyrus (PCG), bilateral dorsal thalamus (DT), bilateral lenticular nucleus (LN), bilateral posterior horn of the lateral ventricle paratrigonal white matter (PWM). In MRS, absolute concentrations of N-acetylaspartate (NAA), myo-inositol (MI), choline-containing compounds (Cho), total creatine (tCr), glutamine + glutamate (Glx) were quantified using LC Model and SAGE software. In DTI, we used fractional anisotropy (FA) and mean diffusivity (MD) to reflect white matter integrity. In Ctx+ patients, scores of functional assessment of cancer treatment cognition test (FACT-Cog), perceived cognitive impairments (PCI), impact of perceived impairments on quality of life (QOL), perceived cognitive abilities (PCA), auditory-verbal learning test (AVLT) recognition and clock drawing test (CDT) were lower than those before chemotherapy (p < 0.05). In MRS, Ctx+ patients had significantly lower NAA values in bilateral PCG, DT, respectively. The concentrations of tCr were observed to decline in bilateral PCG and right DT. Glx values decreased in right DT. Cho values decreased significantly in bilateral DT. In DTI, Ctx+ patients had lower FA values in bilateral PCG compared with patients before chemotherapy. Among imaging metrics and cognitive scores, positive correlations were observed between changes in AVLT recognition scores and changes in NAA values in bilateral PCG (left PCG: r = 0.470, p < 0.01; right PCG: r = 0.500, p < 0.01). Positive correlations were also found between changes in AVLT recognition and changes in FA values in bilateral PCG (left PCG: r = 0.513, p < 0.01; right PCG: r = 0.563, p < 0.01). Chemotherapy can lead to a decrease in memory function, accompanied by changes in brain metabolite concentration and white matter integrity in some parts of brain.

A growing number of studies reported that rosiglitazone (a PPARgamma agonist) could ameliorate the painful state and prevent stress-induced depression. However, whether rosiglitazone can prevent pain-induced depression is unclear. This study aimed to explore the antidepressant effects of rosiglitazone in L5 spinal nerve transection (SNT) induced neuropathic pain rats. In addition, AMPK inhibitor (Compound C) and autophagic antagonist (3-methyladenine, 3-MA) were applied to investigate the underlying therapeutic mechanisms. L5 SNT-induced neuropathic pain symptoms and depressive like-behaviors were detected by paw pressure threshold test (PPT), open-field test (OFT), forced swimming test (FST), tail suspension test (TST), sucrose preference test (SPT). Rosiglitazone could ameliorate L5 SNT-induced neuropathic pain symptoms and depressive like-behaviors and the effect could be reversed by Compound C or 3-MA. Compared with the sham group, the levels of BDNF, AMPK, Beclin-1 and LC3B in rats hippocampus significantly decreased in L5 SNT group. On the contrary, rosiglitazone administration significantly up-regulated the levels of AMPK, BDNF, Beclin-1 and LC3B in rats hippocampus. Compared with sham group, the levels of TNF-α, IL-1β, superoxide dismutase (SOD) and malondialdehyde (MDA) in rat hippocampus significantly increased in L5 SNT group. Besides, rosiglitazone administration significantly decreased the levels of TNF-α, IL-1β, SOD and MDA in hippocampus. Compared with rosiglitazone group, 3-MA administration, but not Compound C administration, significantly increased the levels of TNF-α, IL-1β, SOD and MDA in hippocampus. In conclusion, rosiglitazone can counteract down-regulation of AMPK and BDNF induced by L5 SNT rats in hippocampus, and activate autophagic pathway. These effects may contribute to the antidepressant effect of rosiglitazone on the rats with depression induced by L5 SNT.

Acetoin is an important platform chemical with a variety of applications in foods, cosmetics, chemical synthesis, and especially in the asymmetric synthesis of optically active pharmaceuticals. It is also a useful breath biomarker for early lung cancer diagnosis. In order to enhance production of optical (S)-acetoin and facilitate this building block for a series of chiral pharmaceuticals derivatives, we have developed a systematic approach using in situ-NADH regeneration systems and promising diacetyl reductase. Under optimal conditions, we have obtained 52.9 g L−1 of (S)-acetoin with an enantiomeric purity of 99.5% and a productivity of 6.2 g (L h)−1. The results reported in this study demonstrated that the production of (S)-acetoin could be effectively improved through the engineering of cofactor regeneration with promising diacetyl reductase. The systematic approach developed in this study could also be applied to synthesize other optically active α-hydroxy ketones, which may provide valuable benefits for the study of drug development.