A Tradeoff Between Accuracy and Speed for K-Means Seed Determination

With a sharp increase in the information volume, analyzing and retrieving this vast data volume is much more essential than ever. One of the main techniques that would be beneficial in this regard is called the Clustering method. Clustering aims to classify objects so that all objects within a cluster have similar features while other objects in different clusters are as distinct as possible. One of the most widely used clustering algorithms with the well and approved performance in different applications is the k-means algorithm. The main problem of the k-means algorithm is its performance which can be directly affected by the selection in the primary clusters. Lack of attention to this crucial issue has consequences such as creating empty clusters and decreasing the convergence time. Besides, the selection of appropriate initial seeds can reduce the cluster’s inconsistency. In this paper, we present a new method to determine the initial seeds of the k-mean algorithm to improve the accuracy and decrease the number of iterations of the algorithm. For this purpose, a new method is proposed considering the average distance between objects to determine the initial seeds. Our method attempts to provide a proper tradeoff between the accuracy and speed of the clustering algorithm. The experimental results showed that our proposed approach outperforms the Chithra with 1.7% and 2.1% in terms of clustering accuracy for Wine and Abalone detection data, respectively. Furthermore, achieved results indicate that comparing with the Reverse Nearest Neighbor (RNN) search approach, the proposed method has a higher convergence speed.     

Convolutional Neural Network Auto Encoder Channel Estimation Algorithm in MIMO-OFDM System

Higher transmission rate is one of the technological features of prominently used wireless communication namely Multiple Input Multiple Output-Orthogonal Frequency Division Multiplexing (MIMO–OFDM). One among an effective solution for channel estimation in wireless communication system, specifically in different environments is Deep Learning (DL) method. This research greatly utilizes channel estimator on the basis of Convolutional Neural Network Auto Encoder (CNNAE) classifier for MIMO-OFDM systems. A CNNAE classifier is one among Deep Learning (DL) algorithm, in which video signal is fed as input by allotting significant learnable weights and biases in various aspects/objects for video signal and capable of differentiating from one another. Improved performances are achieved by using CNNAE based channel estimation, in which extension is done for channel selection as well as achieve enhanced performances numerically, when compared with conventional estimators in quite a lot of scenarios. Considering reduction in number of parameters involved and re-usability of weights, CNNAE based channel estimation is quite suitable and properly fits to the video signal. CNNAE classifier weights updation are done with minimized Signal to Noise Ratio (SNR), Bit Error Rate (BER) and Mean Square Error (MSE).     

A Smart Deep Convolutional Neural Network for Real-Time Surface Inspection

A proper detection and classification of defects in steel sheets in real time have become a requirement for manufacturing these products, largely used in many industrial sectors. However, computers used in the production line of small to medium size companies, in general, lack performance to attend real-time inspection with high processing demands. In this paper, a smart deep convolutional neural network for using in real-time surface inspection of steel rolling sheets is proposed. The architecture is based on the state-of-the-art SqueezeNet approach, which was originally developed for usage with autonomous vehicles. The main features of the proposed model are: small size and low computational burden. The model is 10 to 20 times smaller when compared to other networks designed for the same task, and more than 700 times smaller than general networks. Also, the number of floating-point operations for a prediction is about 50 times lower than the ones used for similar tasks. Despite its small size, the proposed model achieved near-perfect accuracy on a public dataset of 1800 images of six types of steel rolling defects.     

智能数据分区与布局研究

大数据时代,数据规模庞大,由数据进行驱动的应用分析场景日益增多.如何快速、高效地从这些海量数据中提取出用以分析决策的信息,给数据库系统带来重大挑战.同时,现代商业分析决策对分析数据的实时性要求数据库系统能够同时快速处理ACID事务和复杂的分析查询.然而,传统的数据分区粒度太粗,且不能适应动态变化的复杂分析负载;传统的数据布局单一,不能应对现代大量增加的混合事务分析应用场景.为了解决以上问题,“智能数据分区与布局”成为当前的研究热点之一,它通过数据挖掘、机器学习等技术抽取工作负载的有效特征,设计最佳的分区策略来避免扫描大量不相关的数据,指导布局结构设计以适应不同类型的工作负载.首先介绍了智能数据分区与布局的相关背景知识,然后对智能数据分区与布局技术的研究动机、发展趋势、关键技术进行详细的阐述.最后,对智能数据分区与布局技术的研究前景做出总结与展望.     

Sn-containing Si3N4-based composites for adaptive excellent friction and wear in a wide temperature range

Ceramic design based on reducing friction and wear-related failures in moving mechanical systems has gained tremendous attention due to increased demands for durability, reliability and energy conservation. However, only few materials can meet these requirements at high temperatures. Here, we designed and prepared a Sn-containing Si₃N₄-based composite, which displayed excellent tribological properties at high temperatures. The results showed that the friction coefficient and wear rate of the composites were reduced to 0.27 and 4.88 × 10^?6 mm³ N^?1 m^?1 in air at 800 °C. The wear mechanism of the sliding pairs at different temperatures was revealed via detailed analyses of the worn surfaces. In addition, the tribo-driven graphitization was detected on the wear surfaces and in the wear debris, and the carbon phase was identified by SEM, TEM, and Raman spectrum.     

Systematic optimization of corrosion,bioactivity, and biocompatibility behaviors of calcium-phosphate plasma electrolytic oxidation (PEO) coatings on titanium substrates

To optimize the corrosion, bioactivity, and biocompatibility behaviors of plasma electrolytic oxidation (PEO) coatings on titanium substrates, the effects of five process variables including frequency, current density, duty cycle, treatment time, and electrolyte Ca/P ratio were evaluated. In our systematic study, a Taguchi design of experimental based on an L16 orthogonal array was used. For this, the coatings characteristics such as the surface roughness, wettability, rutile to anatase and Ca/P ratios, and corrosion polarization resistance were investigated. After determining the optimum process variables for each response, the apatite forming ability in SBF (bioactivity behavior) and MG63 cell attachment and flattening (biocompatibility behavior) for two groups of coatings were examined. The first group was optimized based on the maximum corrosion polarization resistance and the variables were set as the frequency of 2000 Hz, the current density of 5 A/dm², the duty cycle of 30%, the treatment time of 5 min, and the Ca/P ratio of 0.65 at. % in the electrolyte. For the second group, the maximum surface roughness, greatest Ca/P ratio, and highest wettability as well as the minimum rutile to anatase ratio in coatings, could be obtained when the variables were set as the frequency of 10 Hz, the current density of 12.5 A/dm², the duty cycle of 50%, the treatment time of 12.5 min, and the Ca/P ratio of 1.70 at. % in the electrolyte. The results showed that while both groups of coatings indicated a significant apatite forming ability and can serve as bioactive coatings, a proper attachment and flattening of cells and consequently, the favorable biocompatibility properties were seen only in the first group.     

Compatible performance of low thermal conductivity and high infrared radiation of Sm2O3-HfO2 series ceramics applied for thermal protection coatings

Sm₂O₃-HfO₂ series ceramics were synthesized at high temperature using the solid-state reaction. The phase stability, thermo-physical and infrared emission properties of Sm₂Hf₂O₇ (SHO) and Sm₂Hf₂O₇-44.83 wt%HfO₂ (25S/H) composite ceramics were comparatively investigated. Furthermore, their calcium magnesium aluminosilicate (CMAS) corrosion was conducted at 1250°C for different times. The results reveal that both SHO and 25S/H ceramics have excellent phase stability at 1600°C as well as excellent sintering resistance. SHO still exhibits slightly lower thermal conductivity and lower hardness and Young's modulus, higher thermal expansion coefficient (CTE) and fracture toughness as well as higher infrared emittance (0.899 at 800°C) than 25S/H composite with the excessive HfO₂ inside. Both SHO and 25S/H ceramics react with CMAS to form a relatively compact reaction layer, which can effectively prevent the penetration of CMAS. These results preliminarily indicate that SHO ceramic can be proposed as an alternative material of the traditional YSZ for high-temperature thermal protective applications thanks to its compatible performance of low thermal conductivity and high infrared radiation, etc.     

Formation of single-crystalline BaTiO3 nanorods from glycolate by tuning the supersaturation conditions

We have studied peculiarities in the formation of single-crystalline barium titanate (BaTiO₃) nanorods from a glycolate-mediated complex via a single-step hydrothermal process under different supersaturation (S_R) conditions. X-ray diffraction (XRD) showed the formation of pure BaTiO₃ with an S_R of above 19. The tetragonality for the BaTiO₃ (c/a) reached 1.013 at S_R = 19–29 and dropped to 1.010 for S_R = 39. According to the transmission electron microscopy (TEM) and XRD analyses, the rod-shaped particles exhibited single crystallinity and crystal growth along the [001] plane. With scanning electron microscopy (SEM), the morphological evolution from a plate-shaped intermediate precursor (S_R = 6–9) to a rod-shaped product with an aspect ratio of 6–9 (S_R = 19–29), and to non-polar material with an irregular structure (S_R = 39), was observed. The negative slope, linear dependence of the particles’ width and length on the supersaturation level in the range S_R = 19–39 was established for the first time. The replacement of the prevailing crystallization mechanism from in-situ topotactic transformation into dissolution-precipitation above S_R = 19 was observed. It was shown that with a simple regulation of the S_R, the structural and morphological characteristics of the obtained BaTiO₃ nanoparticle can be effectively tuned.     

Adaptation to Endoplasmic Reticulum Stress Enhances Resistance of Oral Cancer Cells to Cisplatin by Up-Regulating Polymerase η and Increasing DNA Repair Efficiency

Endoplasmic reticulum (ER) stress response is an adaptive program to cope with cellular stress that disturbs the function and homeostasis of ER, which commonly occurs during cancer progression to late stage. Late-stage cancers, mostly requiring chemotherapy, often develop treatment resistance. Chemoresistance has been linked to ER stress response; however, most of the evidence has come from studies that correlate the expression of stress markers with poor prognosis or demonstrate proapoptosis by the knockdown of stress-responsive genes. Since ER stress in cancers usually persists and is essentially not induced by genetic manipulations, we used low doses of ER stress inducers at levels that allowed cell adaptation to occur in order to investigate the effect of stress response on chemoresistance. We found that prolonged tolerable ER stress promotes mesenchymal–epithelial transition, slows cell-cycle progression, and delays the S-phase exit. Consequently, cisplatin-induced apoptosis was significantly decreased in stress-adapted cells, implying their acquisition of cisplatin resistance. Molecularly, we found that proliferating cell nuclear antigen (PCNA) ubiquitination and the expression of polymerase η, the main polymerase responsible for translesion synthesis across cisplatin-DNA damage, were up-regulated in ER stress-adaptive cells, and their enhanced cisplatin resistance was abrogated by the knockout of polymerase η. We also found that a fraction of p53 in stress-adapted cells was translocated to the nucleus, and that these cells exhibited a significant decline in the level of cisplatin-DNA damage. Consistently, we showed that the nuclear p53 coincided with strong positivity of glucose-related protein 78 (GRP78) on immunostaining of clinical biopsies, and the cisplatin-based chemotherapy was less effective for patients with high levels of ER stress. Taken together, this study uncovers that adaptation to ER stress enhances DNA repair and damage tolerance, with which stressed cells gain resistance to chemotherapeutics.     

Evaluation of the PSMA-Binding Ligand 212Pb-NG001 in Multicellular Tumour Spheroid and Mouse Models of Prostate Cancer

Radioligand therapy targeting the prostate-specific membrane antigen (PSMA) is rapidly evolving as a promising treatment for metastatic castration-resistant prostate cancer. The PSMA-targeting ligand p-SCN-Bn-TCMC-PSMA (NG001) labelled with ²¹²Pb efficiently targets PSMA-positive cells in vitro and in vivo. The aim of this preclinical study was to evaluate the therapeutic potential of ²¹²Pb-NG001 in multicellular tumour spheroid and mouse models of prostate cancer. The cytotoxic effect of ²¹²Pb-NG001 was tested in human prostate C4-2 spheroids. Biodistribution at various time points and therapeutic effects of different activities of the radioligand were investigated in male athymic nude mice bearing C4-2 tumours, while long-term toxicity was studied in immunocompetent BALB/c mice. The radioligand induced a selective cytotoxic effect in spheroids at activity concentrations of 3–10 kBq/mL. In mice, the radioligand accumulated rapidly in tumours and was retained over 24 h, while it rapidly cleared from nontargeted tissues. Treatment with 0.25, 0.30 or 0.40 MBq of ²¹²Pb-NG001 significantly inhibited tumour growth and improved median survival with therapeutic indexes of 1.5, 2.3 and 2.7, respectively. In BALB/c mice, no signs of long-term radiation toxicity were observed at activities of 0.05 and 0.33 MBq. The obtained results warrant clinical studies to evaluate the biodistribution, therapeutic efficacy and toxicity of ²¹²Pb-NG001.