Influence of polarization direction, incidence angle, and geometry on near-field enhancement in two-layered gold nanowires

The influences of polarization direction, incidence angle, and geometry on near-field enhancements in two-layered gold nanowires （TGNWs） have been investigated by using the vector wave function method. When the polarization direction is perpendicular to the incidence plane, the local field factor （LFF） in TGNW decreases first and then increases with the increase in the incidence angle. The minimum LFF is observed at an incidence angle of 41°. It is found that the increase in the dielectric constant of the inner core leads to a decrease in the LFF. With the increase in the inner core radius, the LFF in TGNW increases first and then decreases, and the maximum LFF is observed at an inner core radius of 27 nm. On the other hand, when the polarization direction is parallel to the incidence plane, the collective motions of the induced electrons are enhanced gradually with the decrease in the incidence angle, and hence the near-field enhancement is increased.

An Enhanced Axial-flux Magnetic-geared Machine with Dual-winding Design for Electric Vehicle Applications

Axial-flux magnetic-geared machine(MGM) is a promising solution for electric vehicle applications for combining the virtues of both axial-flux electric machine and magnetic gear. However, generalized MGMs are limited by the torque density issue, accordingly inapplicable to industrial applications. To solve the abovementioned issue, an improved axial-flux magnetic-geared machine with a dual-winding design is proposed. The key merit of the proposed design is to achieve enhanced torque performance and space utilization with the proposed design, which installs a set of auxiliary winding between modulation rings. With the proposed design, overload protection capability, and fault-tolerant capability can be also achieved, for the proposed machine can work with either the excitation of armature windings or auxiliary windings. The pole-pair, slot combination, and parametric design is studied and optimized by the 3d finite-element method and designed C++ optimization software. Electromagnetic analysis and performance comparison indicate that the proposed machine can achieve a torque enhancement of 68.6% compared to the comparison machine.

Near-Field Scattering Enhancement of Perylene Based Aggregates for Random Lasing

Strong near-field scattering enhancement (NFSE) of polyhedral oligomeric silsesquioxanes(POSS) nanoparticles (NPs) aggregates is found through physical simulation. An aggregation of N,N′-di-[3-(isobutyl polyhedral oligomeric silsesquioxanes) propyl] perylene diimide(DPP) which possesses POSS as scatteres experimentally performs strong NFSE, which confirms the physical simulation results. Moreover, coherent random laser is triggered from the DPP aggregates in carbon disulfide. It is the NFSE of POSS NPs connected to both ends of DPP through covalent bonds and the NFSE of their aggregation thanks to DPP’s aggregation that is responsible for the coherent random laser. So, this work develops a method to improve weak scattering of system through construction of molecules, and opens a road to a variety of novel interdisciplinary investigations, involving molecular designing for disordered photonics.

Recent advances in protein conformation sampling by combining machine learning with molecular simulation

The rapid advancement and broad application of machine learning(ML)have driven a groundbreaking revolution in computational biology.One of the most cutting-edge and important applications of ML is its integration with molecular simulations to improve the sampling efficiency of the vast conformational space of large biomolecules.This review focuses on recent studies that utilize ML-based techniques in the exploration of protein conformational landscape.We first highlight the recent development of ML-aided enhanced sampling methods,including heuristic algorithms and neural networks that are designed to refine the selection of reaction coordinates for the construction of bias potential,or facilitate the exploration of the unsampled region of the energy landscape.Further,we review the development of autoencoder based methods that combine molecular simulations and deep learning to expand the search for protein conformations.Lastly,we discuss the cutting-edge methodologies for the one-shot generation of protein conformations with precise Boltzmann weights.Collectively,this review demonstrates the promising potential of machine learning in revolutionizing our insight into the complex conformational ensembles of proteins.

Enhanced Wolf Pack Algorithm (EWPA) and Dense-kUNet Segmentation for Arterial Calcifications in Mammograms

Breast Arterial Calcification(BAC)is a mammographic decision dissimilar to cancer and commonly observed in elderly women.Thus identifying BAC could provide an expense,and be inaccurate.Recently Deep Learning(DL)methods have been introduced for automatic BAC detection and quantification with increased accuracy.Previously,classification with deep learning had reached higher efficiency,but designing the structure of DL proved to be an extremely challenging task due to overfitting models.It also is not able to capture the patterns and irregularities presented in the images.To solve the overfitting problem,an optimal feature set has been formed by Enhanced Wolf Pack Algorithm(EWPA),and their irregularities are identified by Dense-kUNet segmentation.In this paper,Dense-kUNet for segmentation and optimal feature has been introduced for classification(severe,mild,light)that integrates DenseUNet and kU-Net.Longer bound links exist among adjacent modules,allowing relatively rough data to be sent to the following component and assisting the system in finding higher qualities.The major contribution of the work is to design the best features selected by Enhanced Wolf Pack Algorithm(EWPA),and Modified Support Vector Machine(MSVM)based learning for classification.k-Dense-UNet is introduced which combines the procedure of Dense-UNet and kU-Net for image segmentation.Longer bound associations occur among nearby sections,allowing relatively granular data to be sent to the next subsystem and benefiting the system in recognizing smaller characteristics.The proposed techniques and the performance are tested using several types of analysis techniques 826 filled digitized mammography.The proposed method achieved the highest precision,recall,F-measure,and accuracy of 84.4333%,84.5333%,84.4833%,and 86.8667%when compared to other methods on the Digital Database for Screening Mammography(DDSM).

Enhancing Security in QR Code Technology Using AI: Exploration and Mitigation Strategies

The widespread adoption of QR codes has revolutionized various industries, streamlined transactions and improved inventory management. However, this increased reliance on QR code technology also exposes it to potential security risks that malicious actors can exploit. QR code Phishing, or “Quishing”, is a type of phishing attack that leverages QR codes to deceive individuals into visiting malicious websites or downloading harmful software. These attacks can be particularly effective due to the growing popularity and trust in QR codes. This paper examines the importance of enhancing the security of QR codes through the utilization of artificial intelligence (AI). The abstract investigates the integration of AI methods for identifying and mitigating security threats associated with QR code usage. By assessing the current state of QR code security and evaluating the effectiveness of AI-driven solutions, this research aims to propose comprehensive strategies for strengthening QR code technology’s resilience. The study contributes to discussions on secure data encoding and retrieval, providing valuable insights into the evolving synergy between QR codes and AI for the advancement of secure digital communication.

Whisper intelligibility enhancement based on noise robust feature and SVM

A machine learning based speech enhancement method is proposed to improve the intelligibility of whispered speech. A binary mask estimated by a two-class support vector machine （SVM） classifier is used to synthesize the enhanced whisper. A novel noise robust feature called Gammatone feature cosine coefficients （GFCCs） extracted by an auditory periphery model is derived and used for the binary mask estimation. The intelligibility performance of the proposed method is evaluated and compared with the traditional speech enhancement methods. Objective and subjective evaluation results indicate that the proposed method can effectively improve the intelligibility of whispered speech which is contaminated by noise. Compared with the power subtract algorithm and the log-MMSE algorithm, both of which do not improve the intelligibility in lower signal-to-noise ratio （SNR） environments, the proposed method has good performance in improving the intelligibility of noisy whisper. Additionally, the intelligibility of the enhanced whispered speech using the proposed method also outperforms that of the corresponding unprocessed noisy whispered speech.

Machine learning applications for the prediction of extended length of stay in geriatric hip fracture patients

BACKGROUND Geriatric hip fractures are one of the most common fractures in elderly individuals,and prolonged hospital stays increase the risk of death and complications.Machine learning(ML)has become prevalent in clinical data processing and predictive models.This study aims to develop ML models for predicting extended length of stay(eLOS)among geriatric patients with hip fractures and to identify the associated risk factors.AIM To develop ML models for predicting the eLOS among geriatric patients with hip fractures,identify associated risk factors,and compare the performance of each model.METHODS A retrospective study was conducted at a single orthopaedic trauma centre,enrolling all patients who underwent hip fracture surgery between January 2018 and December 2022.The study collected various patient characteristics,encompassing demographic data,general health status,injury-related data,laboratory examinations,surgery-related data,and length of stay.Features that exhibited significant differences in univariate analysis were integrated into the ML model establishment and subsequently cross-verified.The study compared the performance of the ML models and determined the risk factors for eLOS.RESULTS The study included 763 patients,with 380 experiencing eLOS.Among the models,the decision tree,random forest,and extreme Gradient Boosting models demonstrated the most robust performance.Notably,the artificial neural network model also exhibited impressive results.After cross-validation,the support vector machine and logistic regression models demonstrated superior performance.Predictors for eLOS included delayed surgery,D-dimer level,American Society of Anaesthesiologists(ASA)classification,type of surgery,and sex.CONCLUSION ML proved to be highly accurate in predicting the eLOS for geriatric patients with hip fractures.The identified key risk factors were delayed surgery,D-dimer level,ASA classification,type of surgery,and sex.This valuable information can aid clinicians in allocating resources more efficiently to meet patient demand effectively.

Machine learning potential for Ab Initio phase transitions of zirconia

Zirconia has been extensively used in aerospace,military,biomedical and industrial fields due to its unusual combination of high mechanical,electrical and thermal properties.However,the fundamental and critical phase transition process of zirconia has not been well studied because of its difficult first-order phase transition with formidable energy barrier.Here,we generated a machine learning interatomic potential with ab initio accuracy to discover the mechanism behind all kinds of phase transition of zirconia at ambient pressure.The machine learning potential precisely characterized atomic interactions among all zirconia allotropes and liquid zirconia in a wide temperature range.We realized the challenging reversible first-order monoclinic-tetragonal and cubicliquid phase transition processes with enhanced sampling techniques.From the thermodynamic information,we gave a better understanding of the thermal hysteresis phenomenon in martensitic monoclinic-tetragonal transition.The phase diagram of zirconia from our machine learning potential based molecular dynamics simulations corresponded well with experimental results.

Machine Learning Prediction Models of Optimal Time for Aortic Valve Replacement in Asymptomatic Patients

Currently,the decision of aortic valve replacement surgery time for asymptomatic patients with moderate-to-severe aortic stenosis(AS)is made by healthcare professionals based on the patient’s clinical biometric records.A delay in surgical aortic valve replacement(SAVR)can potentially affect patients’quality of life.By using ML algorithms,this study aims to predict the optimal SAVR timing and determine the enhancement in moderate-to-severe AS patient survival following surgery.This study represents a novel approach that has the potential to improve decision-making and,ultimately,improve patient outcomes.We analyze data from 176 patients with moderate-to-severe aortic stenosis who had undergone or were indicated for SAVR.We divide the data into two groups:those who died within the first year after SAVR and those who survived for more than one year or were still alive at the last follow-up.We then use six different ML algorithms,Support Vector Machine(SVM),Classification and Regression Tree(C and R tree),Generalized Linear(GL),Chi-Square Automatic Interaction Detector(CHAID),Artificial Neural Net-work(ANN),and Linear Regression(LR),to generate predictions for the best timing for SAVR.The results showed that the SVM algorithm is the best model for predicting the optimal timing for SAVR and for predicting the post-surgery survival period.By optimizing the timing of SAVR surgery using the SVM algorithm,we observed a significant improvement in the survival period after SAVR.Our study demonstrates that ML algorithms generate reliable models for predicting the optimal timing of SAVR in asymptomatic patients with moderate-to-severe AS.

从局部到全局的零参考低照度图像增强方法

为解决现有的低照度图像增强方法存在的色彩失真、细节损失以及暗区增强不足和亮区增强过度导致低照度图像增强效果不理想的问题,提出了一种从局部到全局的零参考低照度图像增强方法。采用局部照度增强对低照度图像进行像素级增强,改进了自适应光照映射估计函数,提升了照度调整能力,避免了生成大量的迭代参数,提高了模型的推理速度;采用基于Transformer结构的全局图像调整对局部增强后的图像进行全局调整,解决了亮区照度增强过度的曝光问题和暗区照度增强不足的问题,提升了图像的整体对比度;优化损失函数,对低照度图像特征和增强图像特征进行相似性约束,提升了目标检测精度。实验结果表明,LOL数据集上的客观指标峰值信噪比和结构相似性达到了20.18 dB和0.80,MIT-Adobe FiveK数据集上达到了23.31 dB和0.87,ExDark数据集上增强后图像的目标检测精度提高了7.6%,有效提升了低照度图像可视化质量和目标检测效果。

一种面向机器视觉感知的暗光图像增强网络

低光照等恶劣环境下的目标检测一直都是难点,低光照和多雾因素往往会导致图像出现可视度低、噪声大等情况,严重干扰目标检测的检测精度。针对上述问题,提出了一个面向机器视觉感知的低光图像增强网络MVP-Net,并与YOLOv3目标检测网络整合,构建了端到端的增强检测框架MVP-YOLO。MVP-Net采用了逆映射网络技术,将常规RGB图像转换为伪RAW图像特征空间,并提出了伪ISP增强网络DOISP进行图像增强。MVP-Net旨在发挥RAW图像在目标检测中的潜在优势,同时克服其在直接应用时所面临的限制。模型在多个真实场景暗光数据上取得了优于先前工作效果并且能够适应多种不同架构的检测器。其端到端检测框mAP(50%)指标达到了78.3%,比YOLO检测器提高了1.85%。

时域流形特征增强在数控机床轴承故障诊断中的应用

以数控机床轴承的时域振动信号为研究对象,提出一种基于流形学习的特征增强方法。首先,将采集信号的时间序列进行相空间重构,通过计算子相空间的信息熵来构建信号在特征空间中的表示,并以流形距离作为原始信号来集中不同故障类型的度量。然后,使用等距特征映射算法求取信号在特征空间中同胚的低维流形,其结果可用于对故障类型的分类判别。经实例数据集的验证分析发现,信息熵—等距特征映射变换能够在低维特征空间表达并强化轴承时域信号的故障类型特征,可有效应用于数控机床轴承单一和复合故障场景的设备运行诊断。

基于情感语义增强编解码的神经机器翻译方法

针对目前神经机器翻译模型仅依赖平行语料训练而无法充分挖掘深层语言知识的问题,提出一种基于情感语义增强编解码的神经机器翻译方法,旨在通过引入额外的情感语义,提高模型对语言深层次信息的理解能力。首先,利用word2vec技术获取语料中所有单词的词嵌入,将其输入到一个融合模型中进行训练。该融合模型结合了基于GRU和文档嵌入的机制,以获取单词级别和文档级别的情感语义表征;其次,在情感融合阶段,采用加权公式将单词级别和文档级别的情感语义有机地融合,形成更为综合的情感语义表征;最后,将此表征与上下文语义表征按位相加,以全面引入情感信息,并将其作为输入传递到机器翻译模型的编码器和解码器中。在多个基准数据集上的实验显示,相较于传统的Transformer模型,该方法在IWSLT数据集上性能显著提升,BLEU值增加1.3至1.62。在WMT数据集上也取得良好性能,证实了融合情感语义在机器翻译中的有效性。

基于GEE的中国不同生态系统林火驱动力研究

针对同时对大尺度范围内的不同生态系统林火驱动力研究的难题,提出一种基于谷歌地球引擎(google earth engine,GEE)实现大范围不同生态系统林火驱动力的分析方法。首先基于GEE在线获取中国4个主要不同生态系统林火数据集、Sentinel-2卫星影像和驱动因子等信息,再通过Sentinel-2影像提取的归一化燃烧率差值筛选真实林火点,然后利用随机森林、支持向量机和增强回归树法对林火点分类并评价其表现,最后筛选最佳方法进行林火驱动力重要性分析。研究结果表明:随机森林预测林火的精度最高,均超过92%;山西省长治市和内蒙古大兴安岭地区林火最重要的驱动力分别为人口分布和最高温度,而四川省凉山彝族自治州和江西省赣州市林火发生最重要的两个驱动因子均为帕默尔干旱指数和土壤湿度。研究证明基于GEE的方法可有效地同时实现大范围内中国不同生态系统林火驱动力研究。

区块链技术对企业全要素生产率的影响研究--基于信任构建和降本增效的视角

为了研究区块链技术对企业的生产效率的作用,文章以2019—2021年A股上市公司中非ST制造业企业为样本,运用文本分析及机器学习word2vec方法基于上市公司年报对企业区块链应用程度进行度量,实证检验了其与制造业企业全要素生产率之间的关系。研究发现上市公司区块链技术应用程度越高,其全要素生产率越高。机制研究发现,信任构建和降本增效,是区块链应用赋能全要素生产率提高的主要路径。进一步研究发现,企业子行业类型以及企业代理成本,会影响区块链应用对生产率的促进作用。

EHDE和WHO-SVM模型在齿轮箱故障诊断中的应用

针对现有齿轮箱故障诊断方法对数据长度敏感的缺陷,提出了一种基于增强层次多样性熵(EHDE)和野马算法(WHO)优化支持向量机(SVM)的齿轮箱故障诊断模型。首先,传统熵值特征提取方法在特征提取阶段对数据样本的长度比较敏感,为此提出了增强层次多样性熵,并将其作为特征提取指标用于提取齿轮箱的故障特征;其次,采用WHO算法对SVM模型的参数进行了优化,建立了参数最优的WHO-SVM分类器;最后,将故障特征样本输入至WHO-SVM分类器中进行了训练和识别,完成了样本的故障识别;利用齿轮箱数据集分别从数据长度敏感性、算法特征提取时间、模型诊断性能三种角度对EHDE、精细复合多尺度样本熵、精细复合多尺度模糊熵、精细复合多尺度排列熵、精细复合多尺度散布熵、精细复合多尺度波动散布熵进行了对比研究。研究结果表明:EHDE方法对数据长度的要求较低,在数据长度为512时即可以取得99.1%的平均识别准确率,在诊断稳定性和诊断精度方面均优于其他对比方法;在算法的泛化性实验中,EHDE方法能够以98%的准确率识别齿轮箱的不同故障类型,具有明显的泛化性和通用性。

SERS光谱技术在结直肠癌诊断中的应用效果

选取扬州大学附属江都人民医院收集的50例结直肠癌组血清和50例健康组血清。以金纳米星(AuNS)为基底,利用Renishaw inVia Reflex激光共焦Raman光谱仪,对血清样本的表面增强Raman散射(SERS)光谱进行测定。借助Origin 2021对Min-Max归一化后平均光谱的特征峰分析,并用主成分分析(PCA)-K最邻近(KNN)模型对关键特征提取、分类。结果表明AuNS基底具有良好的均匀性、灵敏度和洁净性。结直肠癌患者在572、633、873、1065、1314、1417和1655 cm^(-1)特征峰处强度明显高于健康组,在1000和1536 cm^(-1)特征峰处强度低于健康组。PCA-KNN模型的准确率达到95%,灵敏度、特异性和曲线下的面积(AUC)分别达到90.0%、96.7%和0.933。结合PCA-KNN模型和SERS光谱技术可实现对结直肠癌快速、准确的识别,为诊断结肠癌提供了一种探索性的新方法。

肠道宏基因组图像增强和深度学习改善代谢性疾病分类预测精度

近年来,统计学和机器学习方法被广泛用于分析人体肠道微生物宏基因组与代谢性疾病之间的关系,这对于微生物群落的功能注释和开发具有重要意义。本研究提出了一种新的可推广的肠道宏基因组图像增强和深度学习框架,用于人类代谢性疾病的分类预测。将3个代表性人类肠道宏基因组数据集中的每个数据样本分别转换为图像并进行数据增强,输入逻辑回归(logistic regression, LR)、支持向量机(support vector machine, SVM)、贝叶斯网络(Bayesiannetwork,BN)和随机森林(randomforest,RF)机器学习模型以及多层感知机(muti-layer perception, MLP)和卷积神经网络(convolutional neural network, CNN)深度学习模型。使用准确率(accuracy, A)、精确率(precession, P)、召回率(recall, R)、F1分数(F1-score)和ROC(receiver operating characteristic)曲线下面积(area under the curve, AUC)5个指标以及10折交叉验证整体评估模型疾病预测的精度性能。结果显示:MLP模型的整体表现优于CNN、LR、SVM、BN、RF以及PopPhy-CNN方法,且经过数据增强(随机旋转和添加椒盐噪声)后,MLP和CNN的模型性能均有进一步提升。MLP模型进行疾病预测的准确率进一步提高了4%~11%,F1提高了1%~6%,AUC提高了5%~10%。以上结果表明,人类肠道宏基因组图像增强和深度学习可以准确地提取微生物群特征,有效预测宿主疾病表型。本研究中使用的源代码和数据集均公开发表在Github中:https://github.com/HuaXWu/GM_ML_Classification.git。

基于改进机器学习的超分辨率图像细节复原

相对于低分辨率图像,高分辨率图像需要增加的像素数目更多,且需要增加高频信息以提升图像的清晰度,当图像目标与背景之间对比度较大时,图像高频细节信息复原难度较高。为此,提出基于改进机器学习的超分辨率图像细节复原方法。对图像去噪,并结合采用双边滤波方法实现图像的对比度增强;利用改进字典的机器学习算法建立双层字典,结合稀疏表示算法获取一层的粗略复原图像;通过二层字典计算一层复原图像与原始图像之间的差值,建立高分辨率样本,并对其开展二层字典训练,通过训练结构实现超分辨率图像的细节复原。实验结果表明,研究方法应用下峰值信噪比可保持在20dB以上,细节复原均方差低于4×10-3,结构相似性指标更高,高分辨率图像的训练效果更好,特征对比明显,细节信息突出。