Adaptive spatial-temporal graph attention network for traffic speed prediction

Considering the nonlinear structure and spatial-temporal correlation of traffic network,and the influence of potential correlation between nodes of traffic network on the spatial features,this paper proposes a traffic speed prediction model based on the combination of graph attention network with self-adaptive adjacency matrix(SAdpGAT)and bidirectional gated recurrent unit(BiGRU).First-ly,the model introduces graph attention network(GAT)to extract the spatial features of real road network and potential road network respectively in spatial dimension.Secondly,the spatial features are input into BiGRU to extract the time series features.Finally,the prediction results of the real road network and the potential road network are connected to generate the final prediction results of the model.The experimental results show that the prediction accuracy of the proposed model is im-proved obviously on METR-LA and PEMS-BAY datasets,which proves the advantages of the pro-posed spatial-temporal model in traffic speed prediction.

Evaluation of right ventricular volume and systolic function in normal fetuses using intelligent spatiotemporal image correlation

BACKGROUND Heart defects are the most common congenital malformations in fetuses.Fetal cardiac structure and function abnormalities lead to changes in ventricular volume.As ventricular volume is an important index for evaluating fetal cardiovascular development,an effective and reliable method for measuring fetal ventricular volume and cardiac function is necessary for accurate ultrasonic diagnosis and effective clinical treatment.The new intelligent spatiotemporal image correlation(iSTIC)technology acquires high-resolution volumetric images.In this study,the iSTIC technique was used to measure right ventricular volume and to evaluate right ventricular systolic function to provide a more accurate and convenient evaluation of fetal heart function.AIM To investigate the value of iSTIC in evaluating right ventricular volume and systolic function in normal fetuses.METHODS Between October 2014 and September 2015,a total of 123 pregnant women received prenatal ultrasound examinations in our hospital.iSTIC technology was used to acquire the entire fetal cardiac volume with off-line analysis using QLAB software.Cardiac systolic and diastolic phases were defined by opening of the atrioventricular valve and the subsequent closure of the atrioventricular valve.The volumetric data of the two phases were measured by manual tracking and summation of multiple slices and recording of the right ventricular end-systolic volume and the right ventricular end-diastolic volume.The data were used to calculate the right stroke volume,the right cardiac output,and the right ejection fraction.The correlations of changes between the above-mentioned indices and gestational age were analyzed.The right ventricular volumes of 30 randomly selected cases were measured twice by the same sonographer,and the intraobserver agreement measurements were calculated.RESULTS Among the 123 normal fetuses,the mean right ventricular end-diastolic volume increased from 0.99±0.34 mL at 22 wk gestation to 3.69±0.36 mL at 35+6 wk gestation.The mean right ventricular end-systolic volume increased from 0.43±0.18 mL at 22 wk gestation to 1.36±0.22 mL at 35+6 wk gestation.The mean right stroke volume increased from 0.62±0.29 mL at 22 wk gestation to 2.33±0.18 mL at 35+6 wk gestation.The mean right cardiac output increased from 92.23±40.67 mL/min at 22 wk gestation to 335.83±32.75 mL/min at 35+6 wk gestation.Right ventricular end-diastolic volume,right ventricular end-systolic volume,right stroke volume,and right cardiac output all increased with gestational age and the correlations were linear(P<0.01).Right ejection fraction had no apparent correlation with gestational age(P>0.05).CONCLUSION Fetal right ventricular volume can be quantitatively measured using iSTIC technology with relative ease and high repeatability.iSTIC technology is expected to provide a new method for clinical evaluation of fetal cardiac function.

Comprehensive modeling of frictional pressure drop during carbon dioxide two-phase flow inside channels using intelligent and conventional methods

Environmentally friendly nature of CO_(2),associated with its safety and high efficiency,has made it a widely used working fluid in heat exchangers.Since CO_(2)has strange thermophysical features,specific models are required to estimate its two-phase characteristics,particularly frictional pressure drop(FPD).Herein,a widespread dataset,comprising 1195 experimental samples for two-phase FPD of CO_(2)was adopted from 10 sources to fulfill this requirement.The literature correlations failed to provide satisfactory precisions and exhibited the average absolute relative errors(AAREs)between 29.29% and 67.69% from the analyzed data.By inspiring the theoretical method of Lockhart and Martinelli,three intelligent FPD models were presented,among which the Gaussian process regression approach surpassed the others with AARE and R^(2)values of 5.48% and 98.80%,respectively in the test stage.A novel simple correlation was also derived based on the least square fitting method,which yielded opportune predictions with AARE of 19.76% for all data.The truthfulness of the newly proposed models was assessed through a variety of statistical and visual analyses,and the results affirmed their high reliability over a broad range of conditions,channel sizes and flow patterns.Furthermore,the novel models performed favorably in describing the physical attitudes corresponding to two-phase FPD of CO_(2).Eventually,the importance of operating factors in controlling the FPD was discussed through a sensitivity analysis.

End-to-End Joint Multi-Object Detection and Tracking for Intelligent Transportation Systems

Environment perception is one of the most critical technology of intelligent transportation systems(ITS).Motion interaction between multiple vehicles in ITS makes it important to perform multi-object tracking(MOT).However,most existing MOT algorithms follow the tracking-by-detection framework,which separates detection and tracking into two independent segments and limit the global efciency.Recently,a few algorithms have combined feature extraction into one network;however,the tracking portion continues to rely on data association,and requires com‑plex post-processing for life cycle management.Those methods do not combine detection and tracking efciently.This paper presents a novel network to realize joint multi-object detection and tracking in an end-to-end manner for ITS,named as global correlation network(GCNet).Unlike most object detection methods,GCNet introduces a global correlation layer for regression of absolute size and coordinates of bounding boxes,instead of ofsetting predictions.The pipeline of detection and tracking in GCNet is conceptually simple,and does not require compli‑cated tracking strategies such as non-maximum suppression and data association.GCNet was evaluated on a multivehicle tracking dataset,UA-DETRAC,demonstrating promising performance compared to state-of-the-art detectors and trackers.

Clinical Intelligent Diagnosis Path Based on the Chief Complaint

Goals of traditional Chinese medicine(TCM)include precision,accuracy,and recognition by clinical practice.Establishment of a diagnosis and treatment system that closely conforms to the principle-method-recipe-medicines system and derivation of an accurate diagnosis and treatment plan should be considerations of TCM.Artificial intelligence research based on computer technology is one of the effective ways to solve this problem.In the research of intelligent diagnosis path,reflecting the characteristics of the overall view and dialectical treatment of TCM such as"Combination of four diagnostic methods""overall examination""combination of disease and syndrome"and"treatment individualized to patient,season and locality"are key for successful research of artificial intelligence in TCM diagnosis or recognition by clinical practice.

Two new approaches for image registration based onspatial-temporal relationship

How to improve the probability of registration and precision of localization is a hard problem, which is desiderated to solve. The two basic approaches (normalized cross-correlation and phase correlation) for image registration are analysed, two improved approaches based on spatial-temporal relationship are presented. This method adds the correlation matrix according to the displacements in x- cirection and y- directions, and the registration pose is searched in the added matrix. The method overcomes the shortcoming that the probability of registration decreasing with area increasing owing to geometric distortion, improves the probability and the robustness of registration.

Multi-dimensional and Multi-threshold Airframe Damage Region Division Method Based on Correlation Optimization

In order to obtain the image of airframe damage region and provide the input data for aircraft intelligent maintenance,a multi-dimensional and multi-threshold airframe damage region division method based on correlation optimization is proposed.On the basis of airframe damage feature analysis,the multi-dimensional feature entropy is defined to realize the full fusion of multiple feature information of the image,and the division method is extended to multi-threshold to refine the damage division and reduce the impact of the damage adjacent region’s morphological changes on the division.Through the correlation parameter optimization algorithm,the problem of low efficiency of multi-dimensional multi-threshold division method is solved.Finally,the proposed method is compared and verified by instances of airframe damage image.The results show that compared with the traditional threshold division method,the damage region divided by the proposed method is complete and accurate,and the boundary is clear and coherent,which can effectively reduce the interference of many factors such as uneven luminance,chromaticity deviation,dirt attachment,image compression,and so on.The correlation optimization algorithm has high efficiency and stable convergence,and can meet the requirements of aircraft intelligent maintenance.

Distributed Intelligent Lighting System by Performing New Model for Illuminance and Color Temperature in the Workplace

A new approach has been proposed to improve the performance of the in-telligent lighting system by estimating personal illuminance and desired color temperature at the workplace. We are considering the problem of using the sensing devices manually for the intelligent lighting system. The lighting control system has not become useful without sensing devices to measure the provided illuminance and color temperature. In this paper, we have used the property of light for the color temperature to estimate the level of color temperature for each user at the workplace. The new method will give personal illuminance for each user at the workplace and decrease the power consumption of the environment as well. As a result, the proposed method of the intelligent lighting system has realized the target of illuminance and color temperature for each user at the workplace by adapting dimming levels using illuminance sensing information for each user. Thus, the energy of the workplace has reduced by using a distributed luminance to realize the target for each user.

Spatial-Temporal Correlation 3D Vehicle Detection and Tracking System with Multiple Surveillance Cameras

Compared to 3D object detection using a single camera,multiple cameras can overcome some limitations on field-of-view,occlusion,and low detection confidence.This study employs multiple surveillance cameras and develops a cooperative 3D object detection and tracking framework by incorporating temporal and spatial information.The framework consists of a 3D vehicle detection model,cooperatively spatial-temporal relation scheme,and heuristic camera constellation method.Specifically,the proposed cross-camera association scheme combines the geometric relationship between multiple cameras and objects in corresponding detections.The spatial-temporal method is designed to associate vehicles between different points of view at a single timestamp and fulfill vehicle tracking in the time aspect.The proposed framework is evaluated based on a synthetic cooperative dataset and shows high reliability,where the cooperative perception can recall more than 66%of the trajectory instead of 11%for single-point sensing.This could contribute to full-range surveillance for intelligent transportation systems.

人工智能定量肺结节参数与肺腺癌浸润程度的相关性分析

目的探讨人工智能(AI)定量肺结节参数与肺腺癌浸润程度的相关性。方法选取2021年1月—2023年1月医院收治的138例肺腺癌患者(共138个肺结节),根据肺腺癌浸润程度将其分为两组,浸润性腺癌(IAC)归为A组(n=60),微浸润性腺癌(MIA)、原位腺癌(AIS)及非典型腺瘤样增生(AAH)归为B组(n=78)。所有患者均行CT扫描,AI肺结节检测系统分析扫描数据,比较两组AI定量肺结节参数,分析AI定量肺结节参数与肺腺癌浸润程度的关系。结果A组肺结节短径、肺结节长径、最大CT值、最小CT值、平均CT值及恶性概率均高于B组(P<0.05);Logistic回归分析显示,各项AI定量肺结节参数升高均是加重肺腺癌浸润程度的独立危险因素(P<0.05);上述AI定量肺结节参数联合预测绘制ROC曲线分析显示联合检测预测准确率最高,其AUC为0.995;Spearman相关性分析显示,AI定量肺结节参数均与肺腺癌浸润程度呈正相关关系(P<0.05)。结论AI定量肺结节参数对浸润性肺腺癌预测价值较高,且与肺腺癌浸润程度具有一定相关性。

时空相关的道路网络短时交通流预测模型

为有效解决复杂路网短时交通流预测问题中涉及的时空特征挖掘问题,提出一种基于改进长短时记忆神经网络(Improved Long Short-Term Memory, ILSTM)的交通流预测模型.首先,通过改进的遗传算法对长短时记忆神经网络(Long Short-Term Memory, LSTM)模型初始参数进行优化获得最优参数组合,解决LSTM初始参数设置对输出结果影响较大的问题.其次,针对复杂路网多路段交通流预测中遇到的空间特征提取问题,通过挖掘相关路段对目标路段交通流预测的影响程度,重新构建LSTM模型的损失函数,采用路网中相关路段对目标路段的影响系数,以损失函数输出值最小为终止条件,构建ILSTM模型.最后,选择加州公路局交通数据进行模型验证实验,采用遗传算法优化LSTM模型(Genetic Algorithm-LSTM, GA-LSTM)和单纯LSTM模型,以及皮尔森相关系数与LSTM组合模型(Pearson Correlation Coefficient-LSTM,PCC-LSTM),对工作日和周末数据的多次实验结果进行对比分析.实验结果表明:ILSTM模型能够充分考虑复杂路网交通流的时间和空间特征,预测平均误差约为1.16%,在收敛效率和预测精度方面均优于其他模型.

融合词性与语义相关性的图书馆智能咨询系统问句相似性计算方法研究

[目的/意义]提高问句相似性计算的准确性,对改善图书馆智能咨询的实际效果具有重要作用。[方法/过程]引入特征词在问句之间的语义相关性指标,提出了融合词性与语义相关性的问句相似性计算方法。首先构建咨询领域专业词典以增强问句处理的业务针对性,然后提出特征词词性权重和语义相关性权重度量方法,并融合两者度量词语在问句中的作用以构建问句语义表征向量,最后采用余弦函数计算问句相似度。[结果/结论]实验表明综合词性与语义相关性特征能有效识别问句中的核心特征词,本文提出的问句相似性计算方法比传统方法具有更高的ACC@1和ACC@3准确率。

基于多种群协同优化的叠合板智能拆分设计

在装配式建筑中,叠合楼板的拆分主要是基于规则的半自动方式完成,其设计过程耗费大量时间和人力,难以获得最优解;同时,采用基于规则的方法存在构件规格多的问题,增加了构件加工的成本和时间。为了解决上述问题,本文提出基于多种群协同优化的叠合楼板智能拆分方法,以提高叠合楼板的标准化和模数化程度。将叠合板规格和数量、接缝宽度规格作为优化目标,每个叠合板位置和尺寸作为优化变量,考虑单构件的限重、限宽等加工和运输要求,建立多目标优化模型。采用变量关联性分析对叠合楼板所有布置区域进行分组划分,原高维优化问题转化为低维优化问题。通过上海某装配式建筑住宅项目的叠合楼板拆分实例应用,表明本文提出的叠合楼板智能拆分方法能够高效、快速地生成全部楼板的规格、数量和位置;相比基于规则的设计方法,本文提出的方法能够一定程度提高叠合板的标准化和模数化水平,且拆分结果满足设计规范和施工要求,从而验证了智能拆分方法的可行性和有效性,有利于装配式建筑的推广。

时序数据分析的复杂化工过程异常智能溯源研究

对复杂化工过程异常工况进行智能推理溯源是实现安全关口前移、降低灾难性事故发生的有效途径。提出了一种基于Spearman-Apriori的化工过程异常智能溯源分析方法,旨在研究复杂化工过程异常工况发生的前置原因,并形成一种智能决策模型。针对化工工艺参数之间耦合性强、关联关系分析难度大的特点,引入Spearman相关系数,通过Spearman实时在线分析过程参数间的相关关系,并设置强关联阈值将Spearman相关系数分析与Apriori算法进行关联耦合,利用Apriori算法中的支持度和置信度二维挖掘各参数之间的超强关联规则。将该方法应用于合成氨工艺中合成工段的异常工况智能推溯,并选取氢氮比、管路工艺气流量、给水换热器冷凝剂流量等8个关键监测指标,研究发现氢氮比增大和给水换热器冷凝剂流量升高分别是导致合成塔入口压力超压、合成塔第一床层温度过低两组异常工况的前置原因,该分析结果与实际生产工艺相符,证明该方法可以有效地对化工过程异常原因进行推溯并筛选主要影响因素。研究为使用生产过程大数据实现化工过程异常智能溯源提供了理论基础,为进一步完善过程风险精细化管控提供了新思路。

基于相关性分析和生成对抗网络的电网缺失数据填补方法

城市电网新型电力系统中多元资源增多,数据采集难度加大,导致数据随机缺失率升高,难以满足精细化分析决策需求。为解决新型电力系统中配网量测数据在采集与传输过程中频发的缺失问题,文中提出一种基于波动互相关分析(fluctuation cross-correlation analysis,FCCA)算法和生成对抗网络(generative adversarial network,GAN)的电网缺失数据填补方法。首先,融合FCCA算法提出强相关性电网数据多维特征提取方法;其次,基于核主成分分析(kernel principal component analysis,KPCA)对多维特征数据集进行降维处理;最后,设计改进型GAN结构,融合电网数据多维特征对低维向量进行重构,实现缺失数据填补。算例采用真实电网数据进行算法验证,并在某城市电网试运行。结果表明,所提方法比传统数据填补方法具有更高填补精度。因此,在新型电力系统中量测数据连续缺失和缺失量较大的情况下,融合强相关性特征进行数据填补,对提升量测数据的完整性和可用性有明显优势。

喉源性咳嗽证型临床特征相关性分析与局部特征AI分析

目的:分析喉源性咳嗽(喉咳)临床症状与中医证型的相关性,探究人工智能(AI)对咽部图像特征分类的分析情况,以期完善临床证候客观信息收集,扩大望诊范围,将精细化局部辨证为完善整体辨证服务。方法:收集2022年8月至2023年2月于上海中医药大学附属岳阳中西医结合医院门诊就诊的喉咳患者142例,记录其相关临床资料及内窥镜下咽后壁图像信息,分析喉咳临床特征与中医证型的相关性,并将所有咽部图像分类标注后交由AI进行局部特征学习,以探究局部特征人工分类的准确率。结果:在单因素Logistic回归模型中,中度咽干(OR=2.630)、中度咽痒(OR=3.290)与痰饮证呈正相关;重度咽干(OR=5.786)、中度咽黏膜充血(OR=7.360)及中度咽后壁淋巴滤泡增生(OR=4.200)与火热证呈正相关;重度咽异物感(OR=5.353)及重度咽黏膜充血(OR=5.194)与瘀血证呈正相关;其余症状与三证型均无明显相关性。将有意义的结果进行多因素Logistic回归分析得出:相对无咽痒,中度咽痒增加痰饮证发生的风险,差异有统计学意义[OR=3.269,95%CI(1.060,10.079),P=0.039);相对于无咽干,中度咽干增加痰饮证发生的风险,差异有统计学意义[OR=2.649,95%CI(1.041,6.745),P=0.041];相对无咽干,重度咽干增加火热证发生的风险,差异有统计学意义[OR=24.615,95%CI(2.774,218.403),P=0.004];相对于无咽黏膜充血,中度咽黏膜充血增加火热证发生的风险,差异有统计学意义[OR=21.819,95%CI(3.526,135.027),P=0.001];相对于轻度咽后壁淋巴滤泡增生,中度咽后壁淋巴滤泡增生增加火热证发生的风险,差异有统计学意义[OR=11.657,95%CI(1.942,69.99),P=0.007];而咽异物感及咽黏膜充血与瘀血证无明显相关性(P>0.05)。AI分析得出此次整体分类的准确率为74.8%。结论:喉咳部分临床特征与证型间具有相关性,可为全身辨证提供参考;其余部分咽喉部特征与全身情况不一定相符,须全身辨证与局部辨证相结合论治;AI图像分析技术与延伸望诊相结合,可拓宽辨证思路,为精细化辨证提供参考。

人工智能驱动的跨模态语义通信系统

概述了跨模态语义通信的相关研究背景,具体包括语义通信面临的两大挑战、跨模态通信的核心思想,以及跨模态语义通信具有的优势与存在的研究空白。针对跨模态语义通信尚存在的研究空白,在人工智能技术的驱动下,提出跨模态语义通信系统架构,详细介绍了跨模态语义通信的核心思想、关键技术,以及实践落地中需要考虑的重要因素,探讨了跨模态语义通信系统的应用场景以及存在的挑战。

基于知识图谱联合ERNIE-DPCNN模型的药品不良反应自动关联性评价方法研究

目的 针对当前药品不良反应关联性评价存在的效率较低和主观性评估问题,通过建立药品不良反应关联性评价模型,探索药品不良反应自动关联性评价方法。方法 利用文献及互联网来源,对获取的不良反应报告标注数据(7 301条)进行知识抽取,构建药品不良反应知识图谱,建立知识驱动的ERNIE-DPCNN自动关联性评价模型。结果 提出的知识图谱联合ERNIE-DPCNN模型在测试集中的精确度、召回率和F1值分别达到82.18%、81.40%、81.21%,相对于其他基线模型各项评估指标均取得了最高值。结论 知识图谱联合ERNIE-DPCNN模型的方法能提高药品不良反应关联性评价效率,具备较强的准确性,并在一定程度上减少主观性评估误差,对基于人工智能的自动化评价有一定参考价值。

岩石变形局部化智能识别的DSCM-CNN方法

岩石变形局部化的识别对于岩石破坏机理、岩土工程灾害预测预警有着重要的意义。本文将数字散斑相关方法(digital speckle correlation methods,DSCM)与卷积神经网络(convolutional neural networks,CNN)相结合,提出了一种用于岩石变形局部化智能识别的DSCM-CNN模型。通过DSCM获取岩石试件在单轴压缩实验过程中的最大剪应变场云图,根据变形局部化带位置进行标注,完成数据集的构建;利用训练数据集对DSCM-CNN智能识别模型进行训练。通过红砂岩单轴压缩实验对该方法进行验证,结果表明:DSCM-CNN模型可以实现岩石变形局部化带位置的自动识别,子集准确率、精确度、召回率等指标分别达到94.19%,97.21%和96.41%,证明了岩石变形局部化智能识别的DSCM-CNN模型的可行性,为岩石变形局部化智能监测提供了一种新的思路。

基于虚拟仿真的高职本科智能制造工程技术实训体系构建

分析了智能制造工程技术实训体系的痛难点,通过融入虚拟仿真实训元素,探索“虚实结合”专业实训教学体系的构建策略。以常州机电职业技术学院智能制造工程技术专业为例,构建实训教学虚实资源支撑专业学生知识、能力和素养目标要求的关联矩阵,形成“依托课程-实体实训-虚拟仿真实训—学生知识能力素养”专业实训资源配置矩阵图,指导实训资源建设、日常实训教学开展,建立目标资源一致的实训教学体系。