ABSTRACT
To address the various challenges in aluminum surface defect detection, such as multiscale intricacies, sensitivity to lighting variations, occlusion, and noise, this study proposes the AluDef-ClassNet model. Firstly, a Gaussian difference pyramid is utilized to capture multiscale image features. Secondly, a self-attention mechanism is introduced to enhance feature representation. Additionally, an improved residual network structure incorporating dilated convolutions is adopted to increase the receptive field, thereby enhancing the network's ability to learn from extensive information. A small-scale dataset of high-quality aluminum surface defect images is acquired using a CCD camera. To better tackle the challenges in surface defect detection, advanced deep learning techniques and data augmentation strategies are employed. To address the difficulty of data labeling, a transfer learning approach based on fine-tuning is utilized, leveraging prior knowledge to enhance the efficiency and accuracy of model training. In dataset testing, our model achieved a classification accuracy of 97.6%, demonstrating significant advantages over other classification models.
ABSTRACT
The Adam algorithm is a common choice for optimizing neural network models. However, its application often brings challenges, such as susceptibility to local optima, overfitting and convergence problems caused by unstable learning rate behavior. In this article, we introduce an enhanced Adam optimization algorithm that integrates Warmup and cosine annealing techniques to alleviate these challenges. By integrating preheating technology into traditional Adam algorithms, we systematically improved the learning rate during the initial training phase, effectively avoiding instability issues. In addition, we adopt a dynamic cosine annealing strategy to adaptively adjust the learning rate, improve local optimization problems and enhance the model's generalization ability. To validate the effectiveness of our proposed method, extensive experiments were conducted on various standard datasets and compared with traditional Adam and other optimization methods. Multiple comparative experiments were conducted using multiple optimization algorithms and the improved algorithm proposed in this paper on multiple datasets. On the MNIST, CIFAR10 and CIFAR100 datasets, the improved algorithm proposed in this paper achieved accuracies of 98.87%, 87.67% and 58.88%, respectively, with significant improvements compared to other algorithms. The experimental results clearly indicate that our joint enhancement of the Adam algorithm has resulted in significant improvements in model convergence speed and generalization performance. These promising results emphasize the potential of our enhanced Adam algorithm in a wide range of deep learning tasks.
ABSTRACT
In this paper, the effect of Ca and Zr additions on microstructure and mechanical properties at room temperature of Mg-Sn alloys was investigated by comparison of Mg-3Sn (wt.%) (T3), Mg-3Sn-1Ca (wt.%) (TX31), and Mg-3Sn-1Ca-1Zr (wt.%) (TXK311) alloys under extrusion. The results show that the main phases of as-extruded T3 alloy were α-Mg and Mg2Sn phases, while the CaMgSn phase was formed and the precipitation of Mg2Sn phase was inhibited in the TX31 and TXK311 alloys due to the addition of the Ca element. Zr did not form intermetallic compounds with other elements but dissolved in the grains of the matrix and became nucleating particles. Incomplete dynamic recrystallization (DRX) occurred in all alloys during hot extrusion. The coarse rod-like and fine block-like mixed CaMgSn phase was observed in α-Mg matrix of as-extruded samples of the TX31 alloy, and the dispersed granular CaMgSn phase was observed in the TXK311 alloy. Ca inhibited the dynamic recrystallization behavior of the alloys, while Zr promoted the dynamic recrystallization behavior. All the as-extruded alloys exhibit typical fiber texture of {0001} basal//ED. With the addition of Ca and Zr elements, the particle stimulated nucleation (PSN) effect excited by the second phase particles gradually weakened the texture. TXK311 alloy has good comprehensive mechanical properties at room temperature, with tensile strength, yield strength, and elongation of 261 MPa, 244 MPa, and 11%, respectively, and the average grain size was 1.8 µm. Grain refinement and second phase dispersion strengthening are considered to play critical roles in the strength optimization of the TXK311 alloy.
ABSTRACT
Fenretinide (4-HPR), a synthetic retinoid, has attracted attention for its anti-inflammation activity. However, few studies have evaluated the effects of 4-HPR on ulcerative colitis (UC). The present study was performed to investigate the therapeutic effects of 4-HPR on UC, and to explore the mechanisms mainly focused on macrophage polarization involved in this progress. Intraperitoneally administered 4-HPR particularly at dose of 100 mg/kg obviously alleviated UC symptoms and restrained the mRNA expression of colonic IL-1ß, IL-6, and TNF-α in dextran sulfate sodium (DSS)-induced mice. Further analysis showed that 4-HPR decreased the mRNA expression of M1 macrophage markers IL-12 and iNOS, while increased M2 macrophage markers Ym1, Arg1 and MRC1 in colonic tissue of mice received DSS. Consistently, an in vitro study revealed that 4-HPR decreased inflammatory response and M1 polarization, while enhanced M2 polarization in LPS-induced RAW264.7 cells. Interestingly, 4-HPR remarkably activated PPAR-γ which was an important regulator of macrophage polarization both in colonic tissue of UC mice and in LPS-induced RAW264.7 cells. Furthermore, these effects of 4-HPR in vivo and in vitro including anti-inflammation and modulation of macrophage polarization were partially abolished by treatment with PPAR-γ antagonist GW9662, indicating that 4-HPR activated PPAR-γ to exert its activities. Taken together, this study demonstrated that 4-HPR might be a potent anti-UC agent that works by regulating macrophage polarization via PPARγ.
Subject(s)
Cell Polarity/drug effects , Colitis, Ulcerative/pathology , Fenretinide/pharmacology , Macrophages/drug effects , Protective Agents/pharmacology , Animals , Colitis, Ulcerative/chemically induced , Colitis, Ulcerative/metabolism , Colon/drug effects , Colon/pathology , Dextran Sulfate/adverse effects , Disease Models, Animal , Male , Mice , Mice, Inbred C57BL , RAW 264.7 CellsABSTRACT
Entangled metallic wire material (EMWM) is a kind of porous damping material. To promote the engineering application of EMWM, it is necessary to establish the constitutive model of EMWM to estimate its mechanical properties. In this paper, a series of quasi-static compression experiments for plate-like EMWM specimens made of austenitic stainless steel wire (06Cr19Ni10) with different densities were carried out in the temperature range of 20-500 °C. It was found that the stiffness of the plate-like EMWM would increase with the increases in the ambient temperature. The non-linear characteristics of the force-displacement curve of the plate-like EMWM would be weakened. Taking the spatial structural characteristics of the plate-like EMWM and the influence of the thermal expansion of the structure into account, a new constitutive model for plate-like EMWM was presented by the combination of the Johnson-Cook model and the Sherwood-Frost constitutive framework model. The accuracy of the model was verified by the experimental data under different temperatures. The results show that the calculated results of the model are consistent with the experimental results. This model can provide an effective theoretical basis for predicting the mechanical properties of plate-like EMWM and guiding its design.
ABSTRACT
BACKGROUND: Depression is commonly observed among patients with gout. Low levels of vitamin D have been associated with depression in non-gout subjects. We examined the association of vitamin D levels with depression in patients with gout. METHODS: We conducted a cross-sectional study of 186 gout patients at the Endocrinology Department of First Affiliated Hospital of Jiaxing University. Levels of serum 25-hydroxyvi-tamin D (25(OH)D) were determined using a competitive protein-binding assay. The 17-item Hamilton Depression Scale was used for screening for depressive symptoms. Diagnosis of depression in gout patients was made in accordance with Diagnostic and Statistical Manual of Mental Disorders, fifth edition criteria for depression. Multivariate analysis was performed using logistic regression models. RESULTS: Thirty-two gout patients (17.2%) were diagnosed as having depression. Patients with depression showed significantly lower 25(OH)D levels as compared to patients without depression (46.4±19.0 vs 57.0±17.3 nmol/L, P<0.001). Significant differences in 25(OH)D quartiles of gout patients were observed between the patients with depression and the patients without depression (P=0.003). In multivariate analyses, serum 25(OH)D levels (≤40.0 nmol/L) were independently associated with depression in patients with gout (OR 3.833, 95% CI 1.406-10.453, P=0.009). CONCLUSION: Our study demonstrates an important association between serum vitamin D levels and depression in patients with gout.
ABSTRACT
OBJECTIVE: To investigate the alternations of thyroid hormone in traumatic patients with severe inflammatory response syndrome (SIRS). METHODS: Fifty traumatic patients with severe SIRS were enrolled and divided into two groups according to whether they presented multiorgan dysfunction syndrome (MODS). Thyroid hormone measurements were taken, including total triiodothyronine (TT3), total thyroxine (TT4), free triiodothyronine (FT3), free thyroxine (FT4) and thyroid stimulating hormone (TSH). The acute physiology and chronic health evaluation II (APACHE II) score was calculated according to clinical data. The outcomes of recovery or deterioration were recorded, as well as the length of time from the onset of SIRS to the time thyroid hormones were measured. RESULTS: Euthyroid sick syndrome (ESS) was presented in 45 cases. TT3 level was negatively correlated with APACHE II score (r = -0.330, P<0.05), and TT3/TT4 value was negatively correlated with the duration of SIRS( r = -0.316, P<0.05). TT3, TT4 and FT3 levels in MODS patients were significantly lower than those without MODS (P<0.05). MODS patients got low TT4 or FT4 level more frequently than those without MODS (P<0.05). Compared with the patients in normal TSH group, the patients with decreased TSH had lower T3, T4, recovery rate and higher APACHE II scores, MODS incidence, but there was no difference between two groups (P>0.05). CONCLUSIONS: Trauma patients with severe SIRS have high possibility to get ESS, which occurs more frequently and severely in MODS patients. It shows the influences of SIRS on the thyroid axes. With the persistence and aggravation of SIRS, there is a progressive reduction of thyroid hormone.
Subject(s)
Euthyroid Sick Syndromes/etiology , Multiple Organ Failure/complications , Systemic Inflammatory Response Syndrome/complications , Wounds and Injuries/complications , APACHE , Adult , Aged , Aged, 80 and over , Female , Humans , Male , Middle Aged , Thyroid Hormones/blood , Wounds and Injuries/bloodABSTRACT
BACKGROUND: The most intimidatory pathological changes in patients with DM are cardiovascular illnesses, which are the major causes of death in diabetic patients and are far more prevalent than in nondiabetics because of accelerated atherosclerosis. In this study, we tried to clarify the changes in macrovascular endothelial ultrastructure and in the gene expression of endothelial nitric oxide synthase (eNOS)mRNA in diabetic rats. METHODS: The study was conducted on 52 of 10-week old Sprague Dawley (SD) rats with body weight of (320 +/- 42) g. SD rats were divided into: experimental group treated with a single intraperitoneal injection of streptozotocin (STZ, 60 mg/kg), (male, n = 20, diabetes mellitus (DMM)); female, n = 12, diabetes mellitus female (DMF)) and control group (male, n = 10, diabetes mellitus male control (DMMC); female, n = 10, diabetes mellitus female control (DMFC)). Four weeks after treatment, half of the rats were sacrificed; the remainders were sacrificed ten weeks after treatment. One part of the abdominal aortic sample was stored under glutaraldehyde (volume fraction psiB = 2.5%). After the process of chemical fixation, chemical dehydration, drying and conductivity enhancement, all samples were observed and photographed using scanning electron microscopy (Leica-Stereoscan 260, England). The other part of the abdominal aortic sample was treated with liquid nitrogen and the expression of eNOSmRNA was assessed by semi-quantitative RT-PCR. RESULTS: The aortic lumen of both experimental groups adsorbed much more debris than that of either control group. The endothelial surfaces of diabetic rats were coarse, wrinkled and protuberant like fingers or villi. The vascular endothelial lesions of diabetic male rats were very distinct after 4 weeks, and as obvious as those at 10 weeks. The vascular endothelial lesions of diabetic female rats were not severe at 4 weeks and only became marked after 10 weeks. In both males and females, the abdominal aortic eNOSmRNA content of 4 weeks and 10 weeks diabetic rats was very significantly lower (P < 0.01) than that of controls. CONCLUSIONS: Aortic endothelial ultrastructure in DM rats is injured compared with controls. Abnormal changes of aortic endothelia in male DM rats are more obvious than those in females. Expression of abdominal aortic eNOSmRNA content of DM rats is significantly lower than that of controls.